What's new in GPU-Z 2.59.0
Apr 19, 2024
- Fixed "unsigned driver cannot load" error on Windows 7/8
- Fixed Resizable BAR status showing "Yes" instead of Enabled/Disabled
- Fixed Resizable BAR status showing as disabled when NVIDIA Optimus GPU is sleeping
- Fixed Resizable BAR status showing as enabled on eGPUs
- Added support for NVIDIA RTX 2000 Ada and RTX 1000 Ada Laptop
New in GPU-Z 2.58.0 (Apr 5, 2024)
- Added initial support for Intel Meteor Lake
- Added support for AMD Radeon RX 7600M
- Added support for NVIDIA RTX 3050 4 GB (GA107), RTX 3500 Ada, A500, A10M, L20, CMP 100-210
- Added VBIOS version reporting for Intel Alder Lake, Raptor Lake, Raptor Lake Refresh
- Added subvendor IDs for Framework and Aetina
- Added support for Shader Model 6.7 and 6.8
- Added support for 8K resolution in Advanced -> DXVA
- The Resizable BAR check will now report the correct capability on NVIDIA Turing
- Improved formatting for some values in Advanced -> OpenCL
- The GPU-Z installer is now digitally signed
- Fixed periodic stuttering on AMD cards in CS2 and other games
- Fixed ROP count on AMD RX 7900 GRE and NVIDIA A40
- Fixed bluescreen on Radeon "Sun" GPU (Southern Islands Oland GPU for mobile, R5 M230/M330)
- Fixed rare bluescreen on some newer AMD integrated GPUs
- Fixed AMD Zen 4 memory type showing as DDR5 when LPDDR5 installed
- Fixed LPDDR5 bandwidth calculation on AMD Zen 4
- Fixed missing VBIOS version readout on AMD integrated GPUs
- Fixed all resolutions sometimes showing "No" in Advanced -> DXVA
New in GPU-Z 2.57.0 (Jan 20, 2024)
- Added support for AMD Radeon RX 7600 XT, 760M (8640H), Steam Deck OLED GPU
- Added support for NVIDIA GeForce RTX 4070 Super, RTX 4070 Ti Super, RTX 4080 Super, RTX 4090D, Tesla T4G, A100 40 GB
- Fixed BIOS saving on some newer NVIDIA cards
- Added detection for AV1 Profile 2 in Advanced -> DXVA
New in GPU-Z 2.56.0 (Nov 29, 2023)
- Added support to list and detect versions of NVIDIA Frame Generation and Ray Reconstruction DLLs in Advanced -> NVIDIA DLSS
- Added "Open Folder" context-menu option in DLSS version scanner
- Added sensor for NVIDIA Crossbar clock (default off)
- Fixed Intel Arc BIOS version reporting
- Fixed release date for Intel Arc A580
- Added support for NVIDIA RTX 5000 Ada, RTX 4500 Ada, RTX 4000 Ada, new RTX 2050 Mobile variant, L40S, H800 PCIe
- Added support for AMD Radeon RX 7900GRE, RX 6750 GRE, Pro W7700, Pro V620 MxGPU
- Added support for Intel Arc A570M, A530M, GPU Flex 170
New in GPU-Z 2.55.0 (Sep 12, 2023)
- Added support for AMD Radeon RX 7800 XT, RX 7700 XT, Ryzen Z1, Radeon Pro W7500
- Added support for NVIDIA GeForce RTX 4060 Ti 16 GB, RTX 3060 3840 SP, RTX 4000/3500/3000 Ada Laptop, RTX A1000 6 GB Laptop, RTX A2000 Embedded
- Intel Arc GPUs now report the overclocked frequency (when active)
- Added transistor count for NVIDIA GK208
- Added die size for NVIDIA GF119 and G98
- Fixed die size for AD107
- Fixed memory bus width detection on some Intel ADL/RPL systems
- Fixed RGB controls not working on Gigabyte RTX 4090 after GPU-Z started
- Fixed ROP count on Navi 33 (RX 7600)
New in GPU-Z 2.54.0 (Jun 21, 2023)
- Added support for NVIDIA GeForce RTX 4060, RTX 4060 Ti, RTX 3060 (GA104-B), RTX 3050 Mobile 4 GB, RTX 3050 Mobile 6 GB (GA107-B), L40-2Q, L4, RTX A500 Mobile, RTX 2000 Ada Mobile, RTX 4000 SFF Ada, RTX 5000 Ada mobile
- Added support for AMD Radeon RX 7600, Pro W7800, W7900, E8860, Ryzen Phoenix Radeon 700M, Ryzen Z1 Extreme
- Added support for Intel Arc Pro A60, A60M, Flex 140, Iris Xe Max 100, additional Raptor Lake iGPUs
- Added support for Sparkle GPU vendor
- Added "Transistors" and "Release Date" to XML dump output
- Added VRAM read/write bandwidth usage sensors on Intel Arc
- Improved codec detection in Advanced -> DXVA (AV1 and others)
- The default tooltip language will now be determined by the logged in user's setting, not the system setting
- Fixed power monitoring on Intel Arc
- Fixed fake detection reporting GF116 when actual GPU is GF119
- Fixed die size for NVIDIA AD102, AD104, AD106, AD107
- Fixed transistor count for RTX 4070 Mobile (AD106)
New in GPU-Z 2.53.0 (Apr 17, 2023)
- Added support for NVIDIA GeForce RTX 4070, Mobile RTX 4090/4080/4070/4060/4050, RTX 6000 Ada, RTX 3060 Mobile (GA104), RTX 3050 Mobile 6 GB, RTX 2050, Hopper H100 PCIe, RTX 2080 Ti ES
- Added support for AMD Radeon RX 7600S, Pro W6900X, Pro V620, Mendocino Ryzen APU
- Added support for Intel Raptor Lake-HX, Alder Lake N, Alder Lake U i3-1210U, Rocket Lake UHD P750
New in GPU-Z 2.52.0 (Dec 22, 2022)
- Added support for AMD Radeon RX 7900 XT and XTX (Navi 31)
- Added support for NVIDIA GeForce RTX 4070 Ti, RTX 3080 Ti 20 GB, RTX 3070 Ti (GA102), RTX 3050 (GA107), A800 80 GB PCIe
- Added support for Intel Raptor Lake IGP
- Added support for AMD Radeon RX 6300
- NVIDIA ECC memory status is now reported in Advanced Panel / General
- If boost clock can't be read, base clocks are used to calculate fillrates
- GPU clock reading on Intel Arc graphics has been improved
- Improved VRAM size reporting in some edge cases
- Added support for vendor detection of Corsair, Huqaqin, Maxsun and Wingtech
New in GPU-Z 2.51.0 (Nov 18, 2022)
- Added full support for NVIDIA GeForce RTX 4080
- Added BIOS save/upload support for NVIDIA GeForce RTX 4080 and 4090
- Fixed wrong transistor count on GeForce RTX 4090
- Added support for monitoring 16-pin power input
- Fixed missing memory temperature on GeForce 40 Series
- Fixed crash in Glenfly Advanced panel
- DLSS Scan in Advanced Panel no longer starts automatically and lets you select the drives to scan first
- When the "Stop" option is selected in DLSS Scan, properly indicate that the search has stopped
- The list of Vulkan extensions is now one entry per line
- The list of OpenCL extensions is now one entry per line and sorted alphabetically
- Fixed negative Gather Offsets range displayed as positive integer in Vulkan info
- Added support for NVIDIA GeForce RTX 3060 Ti GDDR6X, RTX 3060 (GA104-B), MX750 A, RTX A500 Laptop, RTX A4500 Embedded, Tesla T10, Quadro K5100M (GK104-B)
New in GPU-Z 2.50.0 (Oct 1, 2022)
- Added support for NVIDIA GeForce RTX 4090
- Improved detection for Intel Arc A750 & A770
- Added Intel Arc VRAM temperature monitoring
- Fixed Intel Arc temperature sensors
- Instead of showing a blank window, a proper message is displayed when the DLSS Game Scan finds no supported games
New in GPU-Z 2.49.0 (Sep 20, 2022)
- Added support for AMD Ryzen 7000 Zen 4 Raphael integrated graphics
- Fixed Intel Arc A750/780 reporting the wrong GPU
- Added support for Intel Arc A580
- Fixed some rare crashes on AMD cards
- DLSS scan panel in Advanced Tab is now available on non-NVIDIA cards, too
New in GPU-Z 2.48.0 (Sep 7, 2022)
- Added new "DLSS" section to Advanced Tab, which will locate all installed games with DLSS support and report their DLSS version
- GPU-Z will no longer send traffic to www.techpowerup.com and uses www.gpu-z.com exclusively, which makes it easier for IT administrators to block traffic originating from GPU-Z. All previous endpoints on techpowerup.com will be disabled soon, please update your firewall rules accordingly
- When an NVIDIA Engineering Sample GPU is installed, GPU-Z will block all its network activity (feature request by NVIDIA)
- Many improvements to Intel Arc detection, sensors, reporting and specs
- Renamed Intel discrete GPU power sensor to "GPU Chip Power Draw" to clarify that it does not measure whole board power, but GPU chip power only
- Improvements to Chinese translation
- Added detection for Advantech vendor Id
- Fixed fan speed monitoring on Intel DG1 with newer drivers
- Fixed RTX 3080 12 GB release year
- Fixed Ryzen 5800H release date
- Fixed RV670 die size
- Added support for NVIDIA GeForce RTX 3050 OEM, MX550 (TU117-A), RTX A5500, A5500 Mobile, A4500 Mobile, A3000 12 GB Mobile, A1000 Embedded
- Added support for Intel Core i5-1230U, several new Arc SKUs
- Added support for AMD FireStream 9170
New in GPU-Z 2.47.0 (Jul 12, 2022)
- Many improvements to Intel Arc discrete GPU reporting
- Added or improved fake detection for cards based on NVIDIA G98, GT200, GK104
- Added workaround to fix missing clocks on some recent AMD Radeon drivers
- Fixed GeForce RTX 3080 12 GB getting reported as non-LHR
- Fixed specs for NVIDIA GeForce 7800 GS AGP
- Fixed process size, die size and transistor count for AMD Rembrandt
- Fixed AMD logo showing on some ATI cards
- Fixed wrong logo on some Intel Xe IGPs
- Added checkbox in sensor settings to control "Log to file" setting on GPU-Z restart
- Various improvements for Glenfly GPU reporting
- Added Gunnir vendor detection
- Fixed CPU temperature showing as 100°C on unsupported Intel processors
- Added support for NVIDIA GeForce GTX 1630, RTX 3050 Laptop, MX570, A1000, A2000, A3000 12 GB, T400, T500, T550, T600, T1000, T1200
- Added support for Sapphire Radeon RX 6700 non-XT
- Added support for Intel Alder Lake H/U/HX
New in GPU-Z 2.46.0 (May 5, 2022)
- Added support for AMD Radeon RX 6950 XT, RX 6750 XT, RX 6650 XT
- Improved Intel ARC support
- Added support for NVIDIA GeForce RTX 2050 (GA107), NVIDIA A30
- Updated driver to no longer require SSE2 compatible CPU
- Fixed 2022 AMD drivers being reported as "Crimson"
- Fixed Resizable BAR detection on systems with AGP card
- Fixed "email me my validation id" not sending any email
- Added support for iGPU on Alder Lake Mobile
- Added support for Glenfly GPU
New in GPU-Z 2.45.0 (Mar 25, 2022)
- Added support for NVIDIA GeForce RTX 3090 Ti, RTX 3060 Ti (GA103), RTX 3080 Ti Mobile, T1000 8 GB, T400 4 GB, CMP 170HX, A16, A2
- Added support for AMD Radeon RX 6950 XT, RX 6750 XT, RX 6800S, RX 6700S, RX 6850M XT, RX 6650M, RX 6650M XT, W6400, Barcelo APU, Rembrandt APU, Steam Deck GPU, Barco MXRT 270
- Added preliminary support for Intel Arc Alchemist
- Fixed NVIDIA Ampere memory size not being reported as power of two
- Improved support for Intel Alder Lake
- Fixed memory clock detection on Alder Lake
- Fixed GA106 transistor count
- Added support for HBM and DDR4 memory on NVIDIA
New in GPU-Z 2.44.0 (Jan 20, 2022)
- Improved Resizable BAR detection to work better with some edge cases
- BAR sizes are now reported in in Advanced Panel -> Resizable BAR
- GPU-Z will report "Vista 64" as operating system, not "Vista64"
- Screenshots will now be uploaded using https
- Added vendor detection for Vastarmor
- Fixed some GeForce RTX 3060 cards getting labeled as LHR
- Updated release date for AMD Radeon RX 6600
- Added support for NVIDIA GeForce RTX 3050, RTX 3080 12 GB, RTX 3070 Ti Mobile, RTX 3050 Ti Mobile (GA106), RTX 2060 12 GB, GT 1010, MX550, GTX 1650 Mobile (TU117-B), RTX A2000 (GA106-B), RTX A4500, A10G, A100 80 GB PCIe, CMP170HX, CMP70HX
- Added support for AMD Radeon RX 6400, RX 6500 XT, RX 6300M, RX 6500M, W6300M, W6500M, W6600M
- Added support for Intel Alder Lake non-K SKUs, Alder Lake Mobile, Rocket Lake Xeon
New in GPU-Z 2.43.0 (Sep 28, 2021)
- Added ability to read NVIDIA Ampere Laptop power limits in Advanced -> NVIDIA BIOS
- Fixed crash on startup on some older Radeon cards
- Fixed EU count for Intel Rocket Lake
- Fixed screenshot feature crashing under Windows XP. This bug first appeared in 2.39
- Added support for NVIDIA Quadro RTX 3000 (TU106-B)
New in GPU-Z 2.42.0 (Sep 24, 2021)
- Added support for Intel Alder Lake and Tiger Lake Server
- Added display of NVIDIA low hashrate status in GPU name field, e.g. "GA102 (LHR)"
- Added support for new RTX 3060 variant based on GA104
- Added support for resizable BAR detection on Radeon RX 5000 Series
- Added command line parameter "-log", which, when followed by a filename, sets the sensor log filename and begins logging once GPU-Z has started
- Improved EVGA iCX sensor reading stability
- Radeon HD 5000 Series cards will now display the ATI logo
- Fixed AMD Navi 2x not showing as DirectX 12_2 compatible
- Fixed crash during screenshots
- Fixed crash in render test
- Fixed crash on some systems while preparing Resizable BAR report
- Fixed memory clock readings on some AMD APUs
- Added release date for Intel Tiger Lake
- Added support for NVIDIA RTX 3050 Ti Mobile (GA106), T1200 Mobile, GRID K340, GRID M30, Q12U-1
- Added support for AMD Radeon Pro W6800X, Barco MXRT-8700
New in GPU-Z 2.41.0 (Aug 6, 2021)
- Added detection for Windows 11
- Improved TMU estimation for unknown (future) NVIDIA GPUs
- Improved clock speed reporting on professional AMD RDNA2 cards
- The installer will now add the version number to add/remove programs, this improves Winget support
- Always show Navi rated clocks in Advanced Panel, even when some are reported as zero
- Fixed "BIOS reading not supported on this device" on some laptops with NVIDIA dGPU
- Fixed "Lookup" button rendering on ASUS ROG version with non-standard DPI setting
- Updated chinese tooltip translations
- Fixed clockspeed calculation for some older ATI cards (Radeon DDR / 7200 DDR)
- Added die size and transistor count for AMD Cezanne and ATI R100 & RV100
- Added support for AMD Radeon RX 6600 XT, Pro W6800, W6600, Radeon HD 7660G (AMD R-464L APU)
- Added support for NVIDIA CMP 90HX, 50HX, 40HX, 30HX, T1000, T400, A100-SXM-80GB, A10, A5000, A4000, A3000, A2000, RTX 3050 Mobile Series (GA107-B)
New in GPU-Z 2.40.0 (May 28, 2021)
- Added a detailed PCIe Resizable BAR report to Advanced Panel
- Clicking the Resizable BAR status on main screen will open Advanced Panel, PCIe Resizable BAR report
- On AMD, the RDNA2 marketing clocks report in Advanced Panel will now also show rated memory frequency
- Added support for negative temperature monitoring on AMD
- Added support for NVIDIA GeForce RTX 3080 Ti, RTX 3070 Ti, RTX 3080 LHR, RTX 3070 LHR, RTX 3060 Ti LHR, RTX 3050 Mobile, RTX A4000, T600
- Added support for ATI FirePro RG220
New in GPU-Z 2.39.0 (May 10, 2021)
- Added support for Intel Rocket Lake Integrated Graphics
- Added support for NVIDIA RTX 3060 Mobile, RTX 3050 Ti Mobile, RTX 3050 Mobile, RTX A5000, T500, CMP 30HX, CMP 40HX, CMP 90HX
- Added support for AMD Radeon RX 6900 XTXH, Radeon Pro W5500M, Barco MXRT 4700
- Integrated screenshot feature now captures the proper window area on Windows 10 (no longer includes the window shadow)
- VRAM usage sensor removed on NVIDIA cards running in TCC mode, due to unsupported NVIDIA API
- XML dump now includes fields BIOSUEFI, WHQL, DriverDate, DXR, DirectML, OpenGL and ResizableBAR
- Added memory type detection for Intel i740
- Fixed Resizable BAR detection on some systems (mostly laptops)
- Fixed clock readings on AMD Mobile RDNA2
- Fixed OpenCL detection on some rare systems
- Fixed memory clock readings on NVIDIA GeForce 6
- Fixed BIOS date on some legacy ATI cards
- Fixed release date for ATI RV200 and NVIDIA NV41M
New in GPU-Z 2.38.0 (Mar 12, 2021)
- Fixed BIOS saving on AMD RDNA2 cards. BIOSes saved with old GPU-Z versions are only 512 KB and will brick your card when flashed
- PCI-Express Resizable BAR status is now displayed in the main window
- Added support for negative temperature monitoring on NVIDIA
- Added support for AMD Radeon RX 6700 XT, RX 6700, RX 6600 XT, RX 6600, Lucienne APU integrated graphics
- Added support for NVIDIA GeForce RTX 3070 Mobile
New in GPU-Z 2.37.0 (Feb 22, 2021)
- Added memory vendor detection on Navi 1x and Navi 2x
- Added workaround for NVIDIA Ampere PCIe hardware bug
- Added filter to avoid misreadings on EVGA iCX
- Fixed fake detection for some GT218 variants
- Improved Russian translation
- Added preliminary support for Radeon RX 6700 and RX 6600 Series
- Added support for NVIDIA GeForce RTX 3060, RTX 3080 Mobile, RTX 3070 Mobile, RTX 3060 Mobile, RTX A6000, A40, A100-SXM4-40GB, Drive PX2, P106M, Quadro K510M, modded Quadro K6000
- Added support for additional variants of NVIDIA GTX 1650 Max-Q, Quadro P1000, GTX 650, GT 430
- Added support for AMD Cezanne, Radeon Pro V520, R9 290X ES, Barco MXRT 2600
- Added support for Intel Comet Lake Graphics (Celeron 5205U and i7-10810U)
- Added vendor detection for Yeston
New in GPU-Z 2.36.0 (Nov 27, 2020)
- Added support for CPU temperature monitoring on Zen 3
- Improved support for AMD Radeon RX 6800, 6800 XT
- Added BIOS reading on Navi 2x
- Fixed ROP count on Navi 21
- Fixed bluescreen when Windows Code Integrity enabled
- Fixed reported memory size on some NVIDIA cards
- Fixed default clock reading on NVIDIA Ampere when no driver installed
- Fixed IGP detection for Renoir and Intel Xe
- Added support for XFX new vendor ID
- Added support for iCX on EVGA GeForce 30
- Added support for NVIDIA GeForce RTX 3060 Ti, MX450, Quadro RTX 6000 Mobile
- Added support for AMD Radeon Pro VII, HD 8550D (A8-6500T APU), Barco MXRT 5600
- Added support for Intel Xe MAX Graphics, Xe Graphics, Xe Pod, Xe SG-18M Server GPU, Intel Comet Lake iGPU (Xeon W-1250, Xeon W-10855M)
- v2.35.0 (October 15th, 2020)
New in GPU-Z 2.35.0 (Oct 15, 2020)
- Added support for AMD Navi 21 / Radeon RX 6000 Series
- Added support for Intel discrete graphics / DG1
- Added BIOS saving for NVIDIA Ampere
- Fixed memory size reporting on GeForce RTX 3090
- Fixed DirectML detection on Windows 10 Insider 20231.1000
- Improved fake detection for cards based on NVIDIA GT216 & GT218
- Fixed Navi 12 Radeon Pro 5600M detection
- On AMD Cape Verde use direct temperature readings if driver temps are broken
- Added support for NVIDIA A100 PCIe
- Added support for Intel Comet Lake UHD Graphics (Core i5-10200H)
- Added support for AMD Radeon HD 8210E, Barco MXRT-6700
New in GPU-Z 2.34.0 (Sep 7, 2020)
- Added support for NVIDIA Ampere (RTX 3090, RTX 3080, RTX 3070), GeForce GTX 1650 (TU106), GTX 1650 (TU116), Quadro T1000 GDDR6, Tesla V100-SXM2-32GB
- Added support for Intel Tiger Lake, Comet Lake (Xeon W-1290P, Celeron 5205U, Core i3-8140U), Ice Lake (Core i5-1038NG7), Whiskey Lake (Core i5-8265U)
- Added support for Apple Radeon RX 5700 XT & Pro 5700, AMD Radeon RX 350, Radeon E6465
- Added die size for Intel Tiger Lake and Ice Lake
- Added support for DirectX 12 FL 12_2
- Vulkan info in Advanced Panel now lists device extensions and instance extensions, sorted alphabetically
- Fixed memory overclock not showing on Navi when GPU clock at default
- Added support for GDDR6X memory
- Fixed crash during BIOS parsing on Ampere
- Do not calculate NVIDIA driver version number when Microsoft Basic Display Adapter driver active
- Advanced Panel entries will now properly reflow and wrap, when window width is changed
New in GPU-Z 2.33.0 (Jul 3, 2020)
- Fixed D3D12 Mesh Shader detection on Windows 10 2004
- Improved AMD Renoir support, fixed several crashes and issues
- On AMD integrated graphics, report power draw as "GPU Power Draw", not "GPU Chip Power Draw"
- Fixed crash in GPU-Z when "show sensor value in GPU-Z window title" was enabled
- Fixed wrong maximum scale for VRAM usage sensor on some cards
- Fixed some rare cases of AMD memory size getting reported incorrectly
- Fixed crash on startup on older ATI graphics cards
- Added support for AMD Navi 12 / Radeon Pro 5600M
- Added support for NVIDIA RTX 2060 & 2070 Mobile (TU106-B), GeForce MX330 and MX350
- Added support for Comet Lake UHD Graphics 610 & 630, Coffee Lake UHD Graphics P630
New in GPU-Z 2.32.0 (May 29, 2020)
- Added NVIDIA per-rail voltage and power monitoring
- Fixed WDDM 2.7 Hardware Accelerated GPU Scheduling readout in Advanced Tab to report "Disabled, not supported" instead of "Unknown (8)" on Windows 10 2004 with AMD graphics card
- Added support for Intel Comet Lake 10th Generation CPUs
- Added support for AMD Radeon Pro W5700X, WX 4170
- Added support for GTX 1650 Ti Mobile (TU116-A), NVIDIA Tesla V100S-PCIE-32GB
New in GPU-Z 2.31.0 (May 6, 2020)
- Fixed DirectML detection on newer Windows Insider Builds
- Added GPU voltage monitoring for Intel integrated graphics
- AMD Radeon Pro Drivers will now report their version number information
- Added command-line arguments -install and -installSilent
- Switched to InnoSetup based installer
- Improved driver version detection on some systems with NVIDIA GPU
- On Advanced tab, if Vulkan or OpenCL detection fails, report "not supported" instead of "not found"
- GPU-Z startup on slow systems now has longer timeouts to avoid errors
- Added support for NVIDIA GeForce RTX 2070 Super Mobile, RTX 2080 Super Mobile, RTX 2060 Max-Q, RTX 2070 Super Max-Q, RTX 2080 Super Max-Q, RTX 2070 Mobile Refresh, RTX 2060 Mobile Refresh, GTX 1650 Mobile, GTX 1650 Ti Mobile, GeForce MX350, GRID RTX T10 (GeForce Now), Quadro RTX 8000, Tesla P40, Quadro 500M, GeForce GTX 1060 (Microsoft), GeForce GT 610 (GF108), GeForce GT 730M
- Added support for AMD Radeon Pro 580, Radeon Pro V340, Apple 5300M & 5500M
New in GPU-Z 2.30.0 (Mar 13, 2020)
- Added Advanced tab reporting for Hardware Accelerated GPU Scheduling (Windows 10 20H1)
- Advanced tab now shows WDDM 2.7, Shader Model 6.6, DirectX Mesh Shaders, DirectX Raytracing Tier 1.1
- Worked around Microsoft bug to fix DirectML detection on Windows 10 19041 Insider Build
- Driver registry path for the graphics device is now displayed in Advanced -> General
- Renamed NVIDIA "VDDC" sensor to "GPU Voltage"
- Renamed AMD "GPU only Power Draw" sensor to "GPU Chip Power Draw" to clarify that this is the graphics chip only power draw, not the whole graphics card
- Windows Basic Display Driver will no longer show WHQL/Beta status
- Updated Renoir to be 7 nm
- Added support for AMD Radeon RX 590 GME, Radeon Pro W5500, Radeon Pro V7350x2, FirePro 2260, Radeon Instinct MI25 MxGPU, AMD MxGPU
- Added support for Intel UHD Graphics (i5-10210Y)
- Added support for NVIDIA GTS 450 Rev 2
- Fixed crash during DirectX 12 detection
New in GPU-Z 2.29.0 (Jan 15, 2020)
- All AMD RX 5000 Series (Navi) cards now report game clock as "GPU Clock", instead of base clock
- Fixed window position not getting saved when GPU-Z is running during OS shutdown/reboot
- Fixed GPU-Z crash when the AMD driver gets stopped (during driver update etc)
- Fixed PCIe speed reporting on Vega
- Added support for Intel Iris Plus Graphics 645
- Added support for AMD Radeon RX 5600 & 5600 XT, Renoir APU, Radeon Pro Vega II, Radeon HD 8280E
New in GPU-Z 2.28.0 (Dec 9, 2019)
- Added support for AMD Radeon RX 5500 XT, Radeon Pro W5700
- Added support for upcoming AMD graphics driver
- Fixed VRAM reporting on 4 GB Navi cards to show correct value
- Improved detection for AMD RX 470D, RX 560 Mobile, Picasso, Raven Ridge
- Fixed clock readings on older ATI graphics cards
- Added DirectX Support reporting for ATI R200 GPUs
New in GPU-Z 2.27.0 (Nov 18, 2019)
- Added support for AMD Radeon Pro Vega 48, FirePro A300
- Added support for NVIDIA GeForce GTX 1660 Super, GTX 1650 Super, GTX 1650 Max-Q, GeForce 945A, Tesla V100-SXM2-16GB, Tesla P4, Tesla K8
- Advanced Tab now shows PCI bus location of the device
- Fixed crash on very old CPUs due to missing SSE instructions
- Fixed EVGA ICX fan speeds not reported on GeForce 20
- Fixed sensors showing all "0" on some AMD CrossFire setups
- Improved default clock reporting on NVIDIA, even with no driver installed
- Fixed wrong memory clock reported on NVIDIA when two cards with different memory types installed
New in GPU-Z 2.26.0 (Oct 8, 2019)
- Fixed GPU-Z crashing during BIOS save on nearly all recent NVIDIA cards
- Fixed crash on startup on Navi when no driver installed
- Fixed ASUS ROG window getting cut off at bottom, hiding the "Close" button
- Added support for AMD Radeon RX 5500 & RX 5500 Mobile
- Updated the AMD memory temperature tooltip to clarify that it shows memory junction temperature, and that the value is the temperature of the hottest chip (not average)
- Moved reset button on sensors tab by one pixel to properly align it
- Reverted AMD fan-stop-detect-workaround, which was added in last version, because it introduced misreadings. AMD has fixed their driver since 19.9.1
- Fixed PCIe idle speed reported for Vega 20
- Fixed CrossFire detection on Navi and Vega 20
- Fixed NVIDIA Quadro NVS 300 getting reported as fake
- Added fake detection for cards using NVIDIA GK107 & GK208
- Added fake detection for GeForce GTX 1050 Ti based on GK107
- Added support for TU104-based RTX 2060 and Quadro P520
New in GPU-Z 2.25.0 (Sep 11, 2019)
- First tab now shows support status for Vulkan, DirectX Raytracing, OpenGL & DirectML
- Fixed bluescreen in QEMU/KVM virtual machine caused by MSR register access
- Improved clock speed reporting for AMD Navi
- Advanced tab now shows Base, Game and Boost clock on Navi
- Added workaround for stuck fan speeds when fan-stop is active on AMD
- Added workaround for 65535 RPM fan speed reported on Navi
- When BIOS upload is finished show "Finished"
- Added support for NVIDIA Quadro P2200, Quadro RTX 4000 Mobile, Quadro T1000 Mobile
- Added support for AMD Radeon Pro WX 3200, Barco MXRT 7600, 780E Graphics, HD 8330E
- Added support for Intel Ice Lake
New in GPU-Z 2.24.0 (Aug 13, 2019)
- Fixed digital signature error on startup on Windows Vista
- Added support for NVIDIA GeForce 305M, Quadro P620
- Added support for Intel HD Graphics (Xeon E3-1265L V2)
- Fixed some typos in Advanced -> Vulkan
- Added Dataland PCI Vendor ID
New in GPU-Z 2.23.0 (Aug 6, 2019)
- Improved AMD Navi Radeon RX 5700 Series support
- Added support for additional variants of RTX 2060 & RTX 2070 Super
- Updated icon and splash screen with high-DPI support
- Added BIOS saving support for NVIDIA Super cards
- Fixed PCIe speed always getting reported as "4.0" on Navi
- Fixed crashes caused by Valve Anticheat
- Fixed crashes on Radeon R9 290 / 390 Series
- Added support for NVIDIA GeForce GTX 1650 Mobile, GTX 1050 Mobile, GT 1030 (GK107), MX150 (GP107), MX250, Tesla V100-SXM2-16GB, Quadro RTX 5000 Mobile, Quadro RTX 3000
- Added support for AMD Radeon 550X, Radeon 540X, Pro WX 2100, Pro Vega 64, Pro Vega 64X, Pro WX Vega M GL
- Added support for Intel Iris Pro P580, UHD P630, UHD 610 (CFL GT1)
- Added support for ATI Radeon HD 4350 (VisionTek)
- Gainward vendor Id no longer reported as "CardExpert"
- Hardened security of kernel mode driver (CVE-2019-7245)
- When log file can't be written, show error message only once
- Remove readonly attribute from temporary files before trying to delete
- Fixed year in About dialog
New in GPU-Z 2.22.0 (Jul 2, 2019)
- Added preliminary Radeon RX 5700 & 5700 XT (Navi) support
- If a file "GPU-Z.ini" exists in the GPU-Z directory, GPU-Z will use it to read/write all configuration settings instead of the registry, making GPU-Z fully portable
- When an unknown GPU is detected, GPU-Z will show a notice window, which requests a validation submission
- Addded support for PCI-Express Gen 4
- Added support for NVIDIA RTX 2060 Super, RTX 2070 Super, RTX 2080 Super, Tesla T4, Quadro T2000 Mobile
- Added support for AMD FirePro S7150, ATI FireStream 9250
- Added support for Intel HD Graphics 620
- Added AMD SOC Temperature Sensor
New in GPU-Z 2.21.0 (May 24, 2019)
- Fixed missing sensor data on AMD Radeon VII with 19.5.1 drivers
- Fixed broken GPU load sensor on AMD Raven Ridge APUs
- Fix OpenCL detection for Radeon VII and other Vega 20 cards
- Added support for NVIDIA Quadro P500
New in GPU-Z 2.20.0 (Apr 30, 2019)
- Fixed overclock getting reset on NVIDIA cards with Boost
- Fixed crash on Pascal GPUs when no driver loaded
- Fixed crashes on Apple Vega 12
- Changed memory bus width capitalization from "Bit" to "bit"
New in GPU-Z 2.19.0 (Apr 29, 2019)
- Added support for AMD Ryzen 3000 Picasso APUs
- Added support for NVIDIA GeForce GTX 1650, GTX 1650 Mobile, GTX 1660 Ti Mobile,
- GeForce MX250, TU117-B
- Added GeForce MX230 and GP108 transistor count
- AMD Radeon Pro graphics cards now show the appropriate logo
- Improved EVGA iCX to work better and more reliably detect sensor capabilities
- Added support to detect DirectX Raytracing, Variable Rate Shading, WDDM 2.6 and Shader Model 6.5 (listed in Advanced panel)
- Advanced Panel now lists new DX12 capabilities introduced in Windows 10 Updates
- October 2018 and May 2019
- Advanced Panel can now detect Tiled Resources Tier 4
- "ASIC Quality" in Advanced Panel will only be displayed when available, since the feature probably won't be coming back for new GPUs
- Reduced startup time on system with AMD PowerXpress
- Added new icons for the ASUS ROG version and fixed white line below window title
- Fixed crash when no cards detected and hovering over driver info
- Fixed crash on startup on Windows XP machines
- NVIDIA Tesla K80 is GK210, not GK110
New in GPU-Z 2.18.0 (Mar 15, 2019)
- Added support for NVIDIA GeForce GTX 1660, RTX 2060 Mobile (TU106-B), Quadro RTX 5000, Quadro RTX 6000
- Added Hotspot temperature monitoring for AMD Radeon VII
- Vega 10 & Vega 20 Hotspot no longer disabled by default
- Updated PCI vendor database to add new companies and remove unused entries
- Ensure "Sapphire" vendor name doesn't get displayed on PCPartner OEM cards
- Fixed crash on startup during BIOS parsing on older NVIDIA cards (GeForce 500)
- Fixed GPU-Z hang during splash screen on some systems
- Fixed system crash on some AMD systems during BIOS read
New in GPU-Z 2.17.0 (Feb 27, 2019)
- Added support for NVIDIA GTX 1660 Ti, Titan RTX, RTX 2080 Mobile, RTX 2070 Mobile, RTX 2060 Mobile, Quadro RTX 4000, GTX 650 (GK106), Quadro P5200
- Added support for AMD Radeon VII, Radeon HD 8400E
- Added support for Intel Amber Lake GT2 (Core i7-8500Y)
- Added support for detection of Radeon Adrenalin 2019 version
- Simplified some sensor titles: "GPU Clock", "Memory Clock", "Shader Clock"
- Unified "Memory Used" / "Memory Usage" sensor titles to "Memory Used"
- Improved crash reporter to ask for problem description and optional email, so we can follow up
- Advanced Panel dropdown will show "Memory Timings" option only when memory timing details are available
- OpenCL "Max Packet Size" is now formatted as unsigned value
- Instead of empty value show "None" when no OpenCL Built-in Kernels supported
- Added support for "None" in OpenCL DP, SP, Half FP Capability
- Fixed "File creation failed" message during GPU-Z startup
- Fixed GPU and memory load monitoring on RX 580 2048 SP
- Fixed missing default boost clock on GTX 1660 Ti and some Pascal cards
- Fixed missing fan sensors on RTX cards with no monitor connected
- Fixed crash on startup (Windows XP)
- Fixed crash when opening DXVA 2.0 report (Windows XP)
- Fixed power limit reporting on older NVIDIA cards
- Fixed crash during BIOS save on older NVIDIA cards
- Fixed incorrect VRAM reporting on 16 GB Vega
- Fixed various crashes caused by physical memory access
New in GPU-Z 2.16.0 (Dec 11, 2018)
- When a UWD/DCH driver is detected on Windows 10, it will be displayed as "DCH" in driver version
- Advanced tab now includes an entry showing DCH driver status on Windows 10
- Added EVGA iCX support for RTX 2080 FTW3 and RTX 2080 Ti FTW3
- Fixed GPU-Z crash on AMD Polaris
- Added support for NVIDIA GeForce RTX 2060
- Splash screen is now DPI aware
- Fixed sensor list not fully visible on some DPI settings
- Fixed rendering artifacts when resizing sensor window
- Moved "check for updates" setting to main settings panel
New in GPU-Z 2.15.0 (Nov 16, 2018)
- Added hardware giveaway (to the left of "close" button): 2x PowerColor Radeon RX 590 Red Devil, 2x AMD Ryzen 2600X, 15x $30 Steam coupon
- Fixed Quadro K4000 and K2100M detected as fake
- Added support for AMD Radeon RX 590
- Added support for Intel Whiskey Lake, UHD Graphics 617 (Macbook Air 2018)
- Added support for NVIDIA Tesla V100-SXM2-32GB
- Fixed Vega 20 release date
New in GPU-Z 2.14.0 (Oct 25, 2018)
- When available, boost clock is used to calculate fillrate and texture rate
- Fixed missing Intel GPU temperature sensor
- Fixed wrong clocks on some Intel IGP systems ("12750 MHz")
- NVIDIA power sensors now labeled with "W" and "%"
- Added support for Intel Coffee Lake Refresh
New in GPU-Z 2.12.0 (Oct 12, 2018)
- Added detection for fake graphics cards using old relabeled NVIDIA GPUs (G84, G86, G92, G94, G96, GT215, GT216, GT218, GF108, GF106, GF114, GF116, GF119, GK106)
- Added BIOS saving capability for NVIDIA Turing
- Added monitoring for multiple fans on Turing
- Added fan speed % monitoring on Turing
- Added HDMI and DisplayPort info to Advanced -> NVIDIA
- Power draw on NVIDIA cards is now reported in both TDP % and Watt
- Fixed system hang caused by Valve anti-cheat
- Fixed memory bandwidth on Turing with GDDR6
- Fixed tooltip for system memory usage sensor
- Fixed broken Radeon RX 400 GPU usage monitoring on newer drivers
New in GPU-Z 2.11.0 (Sep 17, 2018)
- Added NVIDIA GeForce RTX Turing support
- Added option to minimize GPU-Z on close
- Added system RAM memory usage sensor
- Added temperature monitoring offset for Threadripper 2nd gen
- Fixed typo in NVIDIA Perf Cap Reason tooltip
- GPU-Z will no longer use AMD ADL memory sensors because they are buggy, WDDM monitoring used again
- GPU Lookup feature improved by taking boost clock into account
- Added ability to clean up old QueryExternal files in temp directory
- Added support to BIOS parser for USB-C output, GDDR6 memory, 16 Gbit memory chips
- Added support for NVIDIA RTX 2080 Ti, RTX 2080, RTX 2070, GTX 750 Ti (GM107-A), GTX 1050 Ti Mobile 4 GB, Quadro P1000, Tesla P100 DGXS, GeForce 9200
- Added support for AMD Vega 20, Fenghuang, Ryzen 5 Pro 2500U, 5 Pro 2400G, 3 Pro 2200G, 3 Pro 2300U, 3 2200GE, Athlon 200GE, Embedded V1807B
- Added support for Intel UHD 610, UHD P630 (Xeon), Coffee Lake GT3e (i5-8259U)
New in GPU-Z 2.10.0 (Jul 19, 2018)
- Added CPU temperature monitoring for Intel Sandy Bridge and newer and all AMD Ryzen
- Optimized sensor graph drawing code for faster repaint with lower CPU usage
- Various improvements to avoid error messages related to Query External
- Added button to reset sensor data
- Fixed missing tooltip for GPU clock sensor
- Report Overdrive Limits in Advanced: ATI/AMD BIOS
- Added GDDR6 memory type
- Delay "Screenshot taken" dialog by a second, to not get captured by screenshot
- Fixed secondary devices showing up with no info (old ATI cards)
- Fixed memory clocks on Intel Arrandale
- Fixed memory module info in Advanced Panel for AMD Vega
- Added support for AMD Vega M GH, Vega M GL, WX 4130, WX 4150, WX 8100, Radeon Pro Vega 56, Pro SSG
- Added support for AMD Vega Graphics in Ryzen 3 2200U, Ryzen 3 2300U, Ryzen 3 Pro 2200GE, Ryzen 5 2400GE, Ryzen 7 2400U
- Added support for Intel UHD 605, UHD 610, UHD 630, P580
- Added support for NVIDIA GeForce GTX 1060 6 GB, GTX 1050 3 GB, GTX 1050 2 GB (GP106), GTX 1050 Ti Max-Q, MX110, MX130, GTX 750 Ti (GM107B)
- Added support for NVIDIA Quadro GV100, P600 Mobile, P620, P2000, P3200, P4200, Grid M3-3020
- Added support for NVIDIA Tesla V100, V100-PCIE-32GB, M4, P106-090, P5200, P6, P40, P100 SXM2, P100 PCIe 12 GB
New in GPU-Z 2.9.0 (May 7, 2018)
- Fixed empty or incorrect readings for NVIDIA GPUs in TCC mode (Quadro/Tesla)
- Added WDDM 2.4 support for Windows 10 RS4 / April Update / Spring Creators Update
- Added support for NVIDIA Tesla V100
- Added support for Haswell GT1 in Celeron 2961Y
- Added support for HBM2 memory type in Advanced Panel, NVIDIA BIOS
- Added support for perf cap reason 'none' (Tesla in TCC mode)
- Addded more Bristol Ridge models
- Fixed detection for AMD Carrizo/Bristol/Stoney
- Changed Wani name to Carrizo/Bristol
- Fixed DirectX support reporting for ATI RV200
- Vega's SOC Clock and Hot Spot sensors are now disabled by default at request of AMD. You can enable them any time in settings
- When GPU-Z is set to launch on Window startup, disable "This file was downloaded from Internet" message
- Fixed top right button corruption when resizing the window
- Improved drawing code for sensor graph
New in GPU-Z 2.8.0 (Feb 23, 2018)
- Fixed crashes and other issues on AMD Ryzen Raven Ridge APU
- Added DXVA 2.0 hardware decoder info to Advanced Tab
- "Disable sensor" menu item now properly called "Hide"
- Improved VRAM usage monitoring on AMD
- Improved rendering performance of NVIDIA PerfCap sensor
- Improved AMD power limit reporting in Advanced Panel
- "MemVendor" is now included in XML dump output
- Fixed NVIDIA PerfCap sensor drawing outside its area
- Fixed "BIOS reading not supported" error on NVIDIA, on some motherboards
- Fixed HBM memory type detection in Advanced Tab on Fury X
- Fixed temperature misreadings on Vega
- Fixed "Digital Signature" label getting truncated on some hidpi screens
- Added support for RX 460 Mobile, RX 560 Mobile, RX 570 Mobile, RX 580 Mobile, RX 550 based on Baffin LE
New in GPU-Z 2.6.0 (Jan 16, 2018)
- Sensor refresh will no longer lag the GPU-Z window
- Fixed GPU-Z not starting at all due to broken UPX EXE compressor
- Fixed crashes on RX Vega
- Fixed fan RPM monitoring on RX Vega
- Added support for Radeon Adrenalin Edition
- RX560 GPU renamed from Baffin to Polaris 21 at AMD's request
- Added support for NVIDIA Titan V, GeForce GTX 1060 5 GB, Tesla K40m, GeForce 825M, Quadro M520, NVIDIA NVS 810, NVIDIA Grid M6-8Q
- Added support for AMD Vega 8 Raven Ridge Graphics, Pro WX 9100
- Added support for Intel HD Graphics P630, UHD Graphics 630, Gemini Lake UHD Graphics 600
- Fixed GP100 die size
- Fixed Braswell 16 vs 12 EU count
- Fixed crash on AMD cards with empty board Id (MSI RX 580 Armor)
New in GPU-Z 2.5.0 (Nov 6, 2017)
- Added driver date and WHQL status as separate items
- Added refresh button
- NVIDIA BIOS version is now always uppercased
- Renamed "ForceWare" in driver version to "NVIDIA"
- Updated NVIDIA NVFlash for BIOS saving
- Added support for NVIDIA GTX 1070 Ti, GTX 1050 Ti Mobile (GP106), NVIDIA Quadro GP100, Quadro M620
- Added support for Intel UHD Graphics 620
- Added support for AMD RX Vega 64 Liquid, FirePro M4150
- Added Vega SOC Clock, VR SOC and VR Mem monitoring
- Added support for WDDM 2.3
- Fixed GPU-Z crash when trying to display error message for failed BIOS upload
New in GPU-Z 2.4.0 (Sep 7, 2017)
- Fixed AMD RX Vega shader count detection
- Fixed OpenCl detection on ATI/AMD cards with older drivers
New in GPU-Z 2.3.0 (Sep 5, 2017)
- Added support for AMD Radeon RX Vega Series
- Fixed incorrect/missing information AMD Catalyst 17.2.2 and newer
- Fixed GPU & memory activity monitoring on AMD Radeon RX 500 Series
- Fixed crash on startup in rare CrossFire configurations
- Fixed CrossFire detection for slave cards
- Addded VRM efficiency & VDDC/VDDCI power readings for Polaris
- Added support for AMD Radeon WX 7100, WX 3100
- Addded support for NVIDIA Quadro M2200, GTX 1080 Mobile, GeForce MX150
New in GPU-Z 2.2.0 (Jul 7, 2017)
- Added AMD Radeon memory timings section to Advanced Tab
- Added driver date to driver version tooltip and to Advanced Tab
- Added settings entry to set sensor default display mode (cur, min, max, avg)
- Added right-click menu to Advanced Tab to copy to clipboard
- Improved GPU-Z startup speed, especially on slow or single-core systems
- Query_External files will be deleted from temp folder immediately
- Fixed order of OpenCL entries in Advanced Tab (CMD Queue Properties)
- Fixed missing sensors in shared memory
- Fixed display corruption on older NVIDIA GPUs
- Added EVGA iCX fan monitoring support
- Added die size and transistor count for AMD RX 560
- Added support for Intel Iris Plus 640 & 650, GMA600
- Added support for NVIDIA GeForce GT 1030, GeForce 810M, Mining P104 & P106, Quadro P3000
New in GPU-Z 2.1.0 (May 23, 2017)
New in GPU-Z 2.0.0 (May 23, 2017)
- Advanced Tab added, with additional on info on VGA BIOS, CUDA, OpenCL, DirectX, Vulkan
- Added support for EVGA iCX Sensors
- Fixed BIOS saving on GeForce 900 Series and newer, on 32-bit OS
- NVIDIA PerfCap Reason "Util" renamed to "Idle"
- Added retry option to "Rendering doesn't work over remote desktop" message
- When log file can't be written to, stop trying until re-enabled by user
- Fixed DirectX Support detection on old ATI cards (R600 and older)
- Added support for NVIDIA Tesla P100 PCIe, Tesla M10, Quadro P5000
- Added support for Intel Graphics 615
- Added support for AMD HD 8350G
New in GPU-Z 1.20.0 (Apr 18, 2017)
- Fixed BIOS extraction resulting in corrupted BIOS on Radeon RX 500 Series
New in GPU-Z 1.19.0 (Apr 13, 2017)
- Improved support for AMD Radeon 500 Series
- Added BIOS saving support for NVIDIA GeForce GTX 1080 Ti
- Added support for NVIDIA Titan Xp, Quadro M600M, Quadro M1200
- Fixed NVIDIA driver version detection on Windows 8
- Added new Sapphire vendor Id
New in GPU-Z 1.18.0 (Mar 15, 2017)
- Added support for NVIDIA GeForce GTX 1080 Ti, Quadro P4000, P1000, P600, P400
- Added support for AMD Radeon RX 580, RX 570, RX 560, RX 550, HD 6430M
- Added support for Intel HD Graphics 500 (Apollo Lake Pentium N4200)
- Update check will now display the latest version number available
- Fixed error message display during BIOS upload
- Improved tray icon creation code when GPU-Z is launched during Windows startup
- Fixed clocks read as -1 on Pascal cards with overclocking disabled
- Invalid readings are no longer included in sensor average calculation (avoids -1.$ output)
- Addded support to display Boost clocks on Intel
New in GPU-Z 1.17.0 (Jan 31, 2017)
- Intel GPU codenames now have the CPU name prepended
- All Intel GPUs newer than Sandy Bridge have UEFI marked as available
- Fixed AMD Beema ROP and TMU count, process size 28 nm
- Fixed Bay Trail to be 22 nm, DX11
- Added support for Intel Kaby Lake, Apollo Lake
- Added support for NVIDIA GTX 1050, 1050 Ti Mobile, Quadro 5000, GP104 based GTX 1060, GP104-based Quadro Mobile
New in GPU-Z 1.16.0 (Dec 13, 2016)
- Fixed Polaris OC reading error, introduced with 1.15.0
New in GPU-Z 1.15.0 (Dec 12, 2016)
- Fixed memory clock reading on AMD Fiji cards with Crimson ReLive
- Fixed DDR3 memory detection on Intel Skylake
- Additional fields submitted to GPU Lookup to improve detection
New in GPU-Z 1.14.0 (Dec 9, 2016)
- Fixed clock reading on AMD cards which do not support Wattman
New in GPU-Z 1.13.0 (Dec 9, 2016)
- Added support for Wattman overclocking on older AMD Radeon cards
- Fixed memory leak when running for long time on NVIDIA Pascal
- Fixed driver WHQL detection
- WHQL status is now displayed after the version name
- Fixed reporting format of PCIe bus interface
- Integrated graphics now display N/A as bus interface
- Fixed double error message on startup when log file can't be written to
- Fixed Kaveri DDR3/DDR4 detection
- Added support for AMD Radeon Pro Duo, WX 7100, WX 5100, WX 4100, early support for AMD Stoney
- Added support for VIA graphics
New in GPU-Z 1.12.0 (Oct 18, 2016)
- BIOS upload dialog will now let you browse to the uploaded or duplicate BIOS
- NVIDIA Fermi DirectX support is now listed as 11.0
- GTX 1050 & 1060: SLI support field now shows "Not supported by GPU"
- Sensor window will now scroll by larger distance when clicking outside the thumbtrack
- When GPU-Z window is at the right edge of screen, render test will open to the left of the GPU-Z window
- Added DDR4 detection support for AMD Kaveri
- Added support for NVIDIA GeForce GTX 1080 Mobile, GTX 1070 Mobile, GTX 1060 Mobile, GTX 1050 Ti, GTX 1050, 920 MX (GM108), 940 MX, Quadro P6000
- Added support for AMD R9 M385X, R5 M315
- Added support for Intel HD Graphics 620, HD Graphics 515
New in GPU-Z 1.11.0 (Sep 13, 2016)
- Full set of Device IDs is now displayed in the main window. This can be used to file a compensation claim in the GTX 970 lawsuit.
- Added support for BIOS saving on GTX 1060
- Tray notification icon creation code improved
New in GPU-Z 1.10.0 (Aug 11, 2016)
- Added support for Radeon RX 470, RX 460
- Added support for NVIDIA GTX Titan X Pascal, GTX 1060 3 GB, GT 740 (GK106), 930M (GM108), 930 MX
- Added support for Intel HD Graphics 510, Iris Graphics 550
- Fixed fan-stop on AMD not getting detected correctly
- Fixed NVIDIA driver version not being read on drivers from Windows Update
- Fixed BIOS saving on AMD RX 470/460
- Fixed crash when lookup button clicked when no card detected
- Name for unknown GPUs on ATI/AMD is now displayed correctly
- Fixed incorrect shader count on Polaris GPUs in ULPS mode
- Fixed GTX 1070/1080 release dates
- Clarified sensor name of Polaris to "GPU 'only' power draw"
- Improved power draw monitoring performance on Polaris
New in GPU-Z 1.9.0 (Jul 13, 2016)
- Added support for NVIDIA GeForce GTX 1060
- Fixed BIOS saving on AMD Polaris
- Fixed default clock reading on AMD Polaris
- Added transistor count for AMD Ellesmere
- Added support to indicate SLI high-speed bridge status
- Fixed incorrect sensor readings when AMD GPU is disabled or no driver installed
- Fixed BIOS saving on NVIDIA GTX 690
- HTTPS will be used when connecting to TPU server
- Added support for NVIDIA GT 730 (GK107), GeForce 940MX
New in GPU-Z 0.8.9 (Jun 20, 2016)
- Added AMD Polaris support
- Added support for GeForce GTX 1070
- Added support for reading overclock on GeForce GTX 1070/1080
- Added BIOS reading for GeForce GTX 1070/1080
- Added support for NVIDIA Quadro M2000
- Added support for AMD E8860
- Fixed memory monitoring sometimes resetting beyond 4 GB usage
- Fixed double default memory clock on GT220/240
- Fixed double memory clock on GTxxx cards with GDDR5X
- Fixed typo in BIOS saving dialog
- Fixed no GPU selected in dropdown in some rare cases
New in GPU-Z 0.8.8 (May 13, 2016)
- Added a settings dialog that can be accessed via the icon in the top right of the window
- Added ASIC quality reading for Fiji and Tonga
- ASIC quality readout now compares your GPU against online data from GPU-Z validations
- Added support for NVIDIA GeForce GTX 1080, GTX 980 in laptop, GTX 965M, 940M, 910M, GT 755M, Quadro M6000 24 GB, M4000M, K3100M, M500M, Tesla M40, M60
- Added support for Intel Iris Graphics 540
- Fixed -card parameter not working correctly
- Fixed bluescren on Cedarview, during temperature reading
- Fixed Intel Iris Graphics 540 GPU name to correctly show GT3e
- Fixed NVIDIA BIOS reading on systems with PLX PCIe bridge
- Fixed hang on systems with disabled/broken HPET implementation
- Slighty improved wording in BIOS Upload Dialog
New in GPU-Z 0.8.7 (Feb 23, 2016)
- Radeon Crimson Driver version is now properly detected
- Fixed 1.55V GPU voltage reading on AMD Fiji when in ULPS
- Fixed Skylake DirectX 12 feature level to correctly show 12_1
- Fixed bluescreen on Intel Cloverview (Atom Z2760)
- Fixed CUDA detection for devices on bus number bigger than 9
- Fixed AMD Beema naming
- Fixed OpenCL driver crash message always blaming ATI
- Some HD 2000 and HD 3000 cards are now correctly recognized as ATI
- Revision ID is now always displayed as two digits
- Fixed shader model still being displayed on older cards
- Fixed millisecond precision in log file timestamps
- Updated Armenian language texts
- Misc Stability fixes
- Added support for NVIDIA GTX 980 8GB (notebook), GTX 965M, GTX 750 (GM206), GT 710 (GK208), Quadro K1200, M5000, M2000M, M1000M, K2200M, GRID K160Q, Tesla K80
- Added support for AMD R9 380X, R7 350, Mullins
New in GPU-Z 0.8.6 (Nov 4, 2015)
- Added support for DirectX 12
- Removed shader model readout
- Wake up sleeping AMD GPUs in laptops during GPU-Z startup
- Fixed BIOS reading on AMD Fury Series
- Fixed incorrect fan speed reading on AMD Fury Series
- Fixed Intel Skylake / Braswell / Cherry Lake support
- Fixed several instabilities on Intel GPUs
- Fixed rare bluescreen on AMD Fury Series
- Fixed 3D clocks on AMD not getting read properly when no driver installed
- Fixed sensor misreadings when AMD GPU is in deep sleep
- Fixed rare care of wrong BIOS getting accessed on multi-GPU systems
- Fixed random BIOS version characters when BIOS couldn't be read
- Fixed voltage monitoring on some Sapphire graphics cards
- Memory usage of Intel GPUs is now combined into a single value
- Shortened Biostar vendor name to just "Biostar"
- Improved AMD Radeon memory size detection
- Reduced memory consumption of monitoring sample history
- Removed NVIDIA logo when GPU vendor is unsupported
- Added command line usage with /help
- Added support for NVIDIA GeForce GTX 950, 920M, 845M, GT 710 (GK208), Microsoft Surface Book GPU, Quadro K610M, Quadro M5000M, M4000, M3000M
- Added support for AMD R9 Nano, Carrizo, Kabini
- Added support for Intel Iris Pro P6300, HD Graphics 5600, HD Graphics 520
New in GPU-Z 0.8.5 (Jul 30, 2015)
- Fixed Windows 10 detection
- Fixed CUDA detection in Windows 10
- Added GPU voltage monitoring for AMD Fiji
- Added memory controller load monitoring for AMD Sea Islands, Volcanic Islands, Pirate Islands
- Board Id moved into BIOS version tooltip
- Validation Id now displayed in GPU-Z window title
- Revision Id of recent AMD GPUs will now be displayed
- Added support for NVIDIA GeForce GTX 970M & 980M (GM204-B), GeForce 940M, 930M
- Added support for Intel Atom x7-Z8700, Celeron N3150
- Added support for AMD R9 370, R9 370X, R9 380, R9 390, R9 390X, R9 Fury, A8-7100
New in GPU-Z 0.8.4 (Jun 25, 2015)
- Added PowerColor Radeon Fury X Giveaway
- Added full AMD Radeon Fury X support
- Added support for AMD R7 360
- Added support for Broadwell GT3e
- Added high-quality DPI aware vendor logos
- Whitespace cleanup in Intel VGA BIOS strings
- Fixed UEFI checkbox color on ASUS ROG
- Fixed UEFI checkbox displayed as checked when unknown
New in GPU-Z 0.8.3 (May 28, 2015)
- Added VGA BIOS UEFI support indicator
- Added support for NVIDIA GTX 980 Ti, GTX 965M, GTX 950M, NVS315, GT 750 (GK106)
- Added support for AMD Fiji, Radeon R2/R3 Graphics (Mullins APU)
- Added support for various Intel Broadwell SKUs
- Fixed missing NVIDIA PerfCap sensor in some cases
- Added support for HBM memory
- Fixed GT200B die size
- Fixed some NV42 specs
- Improved OpenCl detection code
New in GPU-Z 0.8.2 (Mar 14, 2015)
- Added graphics card lookup button
- Added Windows 10 support
- Added support for NVIDIA Titan X
- Added support for AMD R9 255, FirePro W7100, HD 8370D, AMD R9 M280X, R9 M295X
- Added support for NVIDIA GTX 980M, GTX 970M, GTX 965M, GTX 845M, GTX 760 Ti OEM, GTX 660 (960 shaders), GT 705, GT 720, GT 745M, NVS 310, Grid K200
- Added support for Intel Broadwell Graphics
- Improved Intel GMA3600 support
- Improved hardware access stability on AMD Kaveri
- Fixed GTX 970 ROP count
- The driver version readout will now display WHQL status
- Added AMD APU logo and updated other logos
- Added PCI-Express bus usage sensor on NVIDIA
- Added registry key (Force_WDDM_Mem_Sensor=1) to bypass NVIDIA driver memory monitoring and use WDDM monitoring instead
- Web links now opened in default browser
- Fixed NVIDIA BIOS save hanging system on Windows 8 and newer
- Show N/A on Intel IGP memory size
- Fixed GPU-Z icon disappearing on shortcuts
- Fixed blank line between sensor logfile entries
- GPU-Z can now start properly from encrypted system drive
- Displayed Catalyst version only includes version number
- Fixed OpenCL detection on HD 7970M
- CrossFire on AMD cards no longer displayed as ATI CrossFire
- Disabled memory vendor detection on future APUs
New in GPU-Z 0.8.1 (Jan 22, 2015)
- Added full GeForce GTX 960 support
- Added preliminary NVIDIA GM200 support
- Fixed NVIDIA GM107 ROP count
- Added support for NVIDIA GTX 980M, GTX 970M, GTX 660 (960 shaders), GT 705, GT 720, GT 745M, NVS 310, Grid K200
- Added support for AMD R9 255, FirePro W7100, HD 8370D
- Added preliminary support for Intel Broadwell Graphics
- Fixed sensor window having wrong size when using startup option -tab 2
- Improved memory vendor detection on NVIDIA
- Added PCI-Express bus usage sensor on NVIDIA
- Fixed OpenCL detection for Vesuvius GPU
- Improved detection for future AMD devices
- Disabled memory vendor detection on future APUs
- Added millisecond precision for sensor logging when refresh faster than 1 sec
- Fixed some rare application crashes
New in GPU-Z 0.8.0 (Oct 23, 2014)
- Added full support for GeForce GTX 970/980
- Fixed BIOS saving on NVIDIA Maxwell GPUs
- Fixed GPU memory usage sensor on Windows 8/8.1
- Fixed system hang when playing video on systems with Hawaii GPU
- Fixed system stuttering on CrossFire with ULPS active
- Added support for AMD Radeon R5 M240, R5 M255, FirePro W2100, W4100, W8100, FireStream 9270, FirePro 2450
- Added support for NVIDIA Quadro K420, K620, K2000D, K2200, K4200, K5200
- Added support for missing Haswell Mobile GPU variant
- Added preliminary support for AMD Topaz GPU
- AMD Tonga: Fix ROP count, shader count, die size, transistors, product name
- Fixed broken HD 7990 detection
- Fixed OpenCL detection on some ATI/AMD cards
- Added support for PerfCap sensor reason: SLI GPUBoost Synchronization
New in GPU-Z 0.7.9 (Aug 12, 2014)
- Added preliminary support for NVIDIA GM204
- Added preliminary support for AMD Tonga
- Added support for AMD Radeon R9 M275X, FirePro W5100, W9100
- Added support NVIDIA GeForce GTX 780 6 GB, GTX 860M, GTX 780M, GT 830M, GT 740, GT 730, GT 720, Quadro NVS 510, FX 380M, GRID K520, Tesla K40c
- Added release date for R9 290
- More robust PhysX detection
- Fixed fan speed monitoring on some recent AMD cards
- Fix for sensor graph over/underflow
- Performance improvements to sensor graph drawing
- Fix for French translation
New in GPU-Z 0.7.8 (Apr 14, 2014)
- GPU-Z window can be resized while sensor tab is active
- Fixed incorrect readings on GeForce Tesla cards
- Added support for AMD Radeon R9 295X2, R9 M275, HD 7500G, FirePro W9000
- Added support for NVIDIA GeForce Titan Z, GTX 745, GTX 880M, 870M, 860M, 850M, 775M, NVS 510
- Fixed release date for GTX Titan Black, GTX 780 Ti
New in GPU-Z 0.7.7 (Feb 14, 2014)
- Sensor refresh rate is adjustable now
- Fixed GM107 TMU & shader count
- Improved NVIDIA Maxwell support
- Added CHiL8214 voltage monitoring for Pitcairn/Curacao cards
- Added support for NVIDIA Tesla K20Xm, GTX 860M
- Added support for AMD Radeon HD 8210
New in GPU-Z 0.7.6 (Jan 27, 2014)
- Added graphics memory vendor detection
- Updated integrated NVFlash for BIOS readings
- Improved voltage monitoring support for GTX 780 Ti
- Added support for new Intel Iris 5200 variant
- Added support for AMD Radeon R7 M265, R5 M230, Kaveri A10-7850K, A10-770K, AMD FireStream 9350, HD 8530M, HD 8650D
- Added support for NVIDIA GTX Titan Black, GTX 750 (GM107), GTX 750 Ti (GM107), GeForce 840M (GM108), GTX 760 (192-bit), GT 750M (Apple), GT 735M, GT 720M
New in GPU-Z 0.7.5 (Dec 20, 2013)
- Fixed crash/hang on NVIDIA Optimus
- Fixed crash during CUDA detection
- Added voltage monitoring support for new Bonaire/Pitcairn SKUs
- Fixed BIOS reading on AMD Mars and Oland
- Added support for AMD Radeon R7 260, HD 7600A, HD 8850M, HD 8400, HD 8240, HD 7620G
- Added support for NVIDIA GeForce 705A, GTX 645, GTX 650 OEM, Grid K260Q, Quadro K5100M, K6000
- Added support for Intel GMA (Bay Trail), HD 4200
- Windows 8.1 will now be detected properly
- Sensor data in shared memory will now update once per second, as intended
New in GPU-Z 0.7.4 (Oct 31, 2013)
- Added support for AMD R9 290X, R9 290, R9 270, HD 7310, HD 8280
- Added support for NVIDIA GTX 780 Ti, GT 635, Quadro K3100M
- Added release date for AMD R7 260X, R7 250, R7 240
- Fixed release date for AMD R9 280X
- Fixed die size for AMD Tahiti
- Fixed ROP count on Ivy Bridge and Haswell
- Fixed BIOS saving not working on AMD cards without driver
- Fixed some rare crashes on systems with Intel VGA
- Render Test can be paused by left-clicking into window
New in GPU-Z 0.7.3 (Sep 13, 2013)
- Added preliminary support for AMD Radeon R7 240, R7 250, R7 260X, R9 270, R9 270X, R9 290, R9 290X
- Added support for AMD Radeon HD 7600G, HD 8250, HD 8330, HD 8470G, HD 8570, HD 8570G, E6460
- Added support for AMD FirePro W7000, W600
- Added support for Intel B43 variant
- Added support for NVIDIA GeForce GT 625, GT 645, GT 755M, GT 745M, GTX 760M, GTX 765M, GTX 770M, Quadro K1100M, K5100M
- Fixed release date for GeForce GT 750M
- Fixed die size for GF117
- Added voltage monitoring support for NVIDIA cards with NCP4206 voltage controller
- Added voltage monitoring support for GeForce GTX 780 with CHL8318
- Fixed crash on AMD HD 7790 while playing video
- Fixed rare issue with fan RPM monitoring when no fan present
- Fixed upgrade installer not working properly when GPU-Z is running
- Sensor history will now be cleared on resume from standby/hibernate
New in GPU-Z 0.7.2 (Jun 21, 2013)
- Added support for NVIDIA GeForce GTX 760, Tesla K10, GT740M (GK208)
- Added support for AMD Radeon HD 8490, HD 8970M, HD 7400D
- Improved support for Intel Haswell graphics
- Added die size, transistors, release date for Haswell
- Added AMD graphics card logo
- Added translations: Greek, Portuguese (Brazil)
- Updated translation: French
- Fixed rare crash during DirectCompute detection
New in GPU-Z 0.7.1 (May 18, 2013)
- Added support for NVIDIA GeForce GTX 770, GTX 780, GTX 650 Ti OEM, GT 640 (GK208), GT 630 (GK208), GT 730M, GT 750M, GTX 780M, GRID K1, GRID K2
- Added support for AMD Radeon HD 7730, HD 8310G, HD 6480G, HD 8410G, HD 8450G, HD 8510G, HD 8550G, HD 8610G, HD 8650G
- Fixed BIOS reading on GeForce GTX 650 Ti Boost
- Added experimental sensor readout to show reason for power capping on recent NVIDIA cards
- Fixed clock readings on some older NVIDIA cards
- Improved detection for future NVIDIA cards
New in GPU-Z 0.7.0 (Apr 18, 2013)
- Added support for AMD Radeon HD 7990, HD 8550M, HD 7340, HD 7290, HD 8670D
- Added support for NVIDIA GeForce GT 740M, 680MX, 675MX, GT218 based 9400 GT
- Fixed TMU count for RV620
- Fixed voltage monitoring on HD 7790
New in GPU-Z 0.6.9 (Mar 19, 2013)
- Fixed shader count on AMD Radeon HD 7790
- Added support for AMD Radeon HD 8870M
- Added support for NVIDIA GeForce GTX 650 Ti Boost, GT 415, GT 750M
- Changed NVIDIA Kepler DirectX support to 11.0
- Fixed OpenCL detection showing double error message
- Added tooltips for GPU vendor logo, save bios, render test, screenshot button
New in GPU-Z 0.6.8 (Mar 7, 2013)
- Added support for GeForce GTX Titan, GeForce 505 (GT216), GT 640M, GTX 675MX, Quadro K2000, K3000
- Added support for AMD Radeon HD 7480D, FirePro W8000, V8700, fake HD 6850 (based on HD 6450)
- Fixed GT 620 GPU name
- Fixed release date for Trinity HD 7000G
- Added detection for Intel GMA
- Improved GPU load monitoring on AMD cards
- Fixed two application crashes related to Intel cards
New in GPU-Z 0.6.7 (Jan 21, 2013)
- Added readout for texture mapping units (TMUs)
- Window position is now restored on startup
- When GPU-Z is minimized, the tray icon tooltip will show some sensor data
- Added support for AMD Tahiti LE, HD 7870M, HD 8000M Series
- Added support for NVIDIA GTX 670MX, GT 645M, 730M, GeForce 505, Tesla K20c, K20m, M2070-Q, M2075, Quadro 410, K600, K4000, K4000M K5000, K5000M 5010M
- Fixed system hang when GPU-Z starts on CrossFire systems (Catalyst driver bug)
- Fixed application crash when scanning for power sensors (NVIDIA driver bug)
- Fixed memory size and GPU temperature reading on AMD Llano Fusion systems
- Added command-line parameter -tab to select active tab on startup
- Fixed OpenCL detection for all AMD Thames based cards (e.g. FirePro m2000)
- Added version number to about dialog
New in GPU-Z 0.6.6 (Oct 26, 2012)
- Fixed GPU-Z window getting cut off on non-standard DPI settings
- Fixed overclock calculation error on NVIDIA non-Kepler GPUs
- Fixed GPU-Z not starting or crashing the machine on systems with broken HPET implementation
- Fixed SLI not getting detected under Windows XP
- Fixed garbled board name on some older Radeon cards
- Fixed several memory leaks
- L6788A voltage controller on HD 7750 and HD 7770 will now display correct default voltage
- Added support for NVIDIA Tesla M2070
New in GPU-Z 0.6.5 (Oct 11, 2012)
- Added BIOS save and upload for NVIDIA Kepler GPUs
- Fixed boost clock detection for NVIDIA Kepler
- Fixed realtime memory clock monitoring for NVIDIA Kepler cards without Boost
- Fixed boost clock readout showing for Kepler cards without Boost
- Fixed clock, temperature readings and crashes on AMD Trinity
- Fixed uninstaller icon not displaying properly
- Additional fixes for window getting cut-off on some DPI settings
- Added working ASIC quality readings for NVIDIA Kepler GPUs
- Added support for Intel Haswell Graphics
- Added support for AMD Radeon HD 7450A, HD 7730M, HD 7700M, FirePro M4000, W5000
- Added support for NVIDIA GeForce GTX 650 Ti, GT 635M, GT 620M, GeForce 610M, GT 620 (GF119), GT 640 (GF118), Quadro 7000, Quadro K2000M
- Added release date for GeForce GTX 660 and GTX 650
- Added die size for NV10, NV11, NV15, NV17, NV18, NV25, NV28, NV30, NV31, NV34, NV35, NV36, NV38, NV42, G70M, G71M, G73M
New in GPU-Z 0.6.4 (Aug 10, 2012)
- Changed sort order of devices to put devices with connected monitor first
- Fixed memory size reading for NVIDIA cards with 4 GB or more
- Fixed AMD Trinity Fusion clock readings
- Added support for NVIDIA GeForce GTX 660 Ti, GTX 660, GTX 650, GT 610, GTX 680M
- Added support for AMD FirePro 2270
- Fixed ROP and TMU count for Intel GMA 3600
- Changed sensor label from "VReg Temperature" to "VRM Temperature"
- Fix for sensors displaying Fahrenheit not being saved properly
- Added VRM temperature monitoring for CHL8228 based cards
- Fix for various crashes
New in GPU-Z 0.6.3 (Jul 16, 2012)
- Added support for NVIDIA GT 620, GT 630, GT 640, GTX 660, Tesla M2090, Quadro 1100M, Quadro 5000M, NVS 5400M, G610M, GT 620M, GT 630M, GT640M LE, GT 660M
- Added support for AMD Trinity, HD 7570, HD 7670, HD 6610M, HD 7550M, HD 7850M, HD 7520G, HD 7640G
- Added warning when OpenCL detection crashed (usually caused by buggy Intel Ivy Bridge GPU drivers)
- Fixed OpenCL detection for Intel, AMD Cape Verde
- Improved NVIDIA Kepler boost clock detection
- Shader clock readout will now be "N/A" when not available
- Bus interface for Fusion and Intel SB/IB will now be "N/A"
- Fixed crash on ATI related to BIOS reading
- Added GPU Power consumption monitoring for Sandy Bridge & Ivy Bridge
- Added voltage monitoring for HD 7700 Series
- Fixed crashes on Windows 8
- Installer will not be launched when GPU-Z is started using -dump
- Added Icon and Publisher name to Installer
- Sensor log checkbox will uncheck properly
New in GPU-Z 0.6.2 (May 3, 2012)
- Fixed boost clocks being displayed on Fermi
New in GPU-Z 0.6.1 (May 3, 2012)
- Added PowerColor giveaway
- Added support for NVIDIA GeForce GTX 690, GTX 670, GT 640, GT 630, GeForce 605, Quadro 3000M, GTX 675M, GTX 670M
- Added support for AMD Radeon HD 7970M, HD 7450
- Added full support for Intel Sandy Bridge and Ivy Bridge integrated graphics
- Fixed ROP count on GK107
- Improved NVIDIA Kepler Boost Clock detection
- Fixed crash on AMD Southern Islands when no graphics card driver installed
- Fixed installer not working properly on Windows XP
- Disabled broken BIOS save on NVIDIA Kepler
- BIOS files are saved as .rom now
- Fixed sensor graph redraw and improved its performance
- 0.6.0
New in GPU-Z 0.6.0 (Mar 20, 2012)
- Added optional GPU-Z installer which launches when GPU-Z is started for the first time, or can be started manually via the GPU-Z system menu
- Added support for AMD Radeon HD 7870, HD 7850, HD 7670M, FireStream 9370, FirePro V3900
- Added support for NVIDIA GeForce GTX 680, GT 640M, GTX 560 SE, GT 550M, GeForce 510, NVS 2100M
- Fixed clocks not showing correctly on some systems with Catalyst 12.2
- Fixed board id reading on pre-Fermi cards
- Fixed OpenCL detection for ATI Fusion
- Fixed GPU-Z window getting cut off at certain DPI settings
New in GPU-Z 0.5.9 (Feb 13, 2012)
- Added support for AMD Radeon HD 7750 and HD 7770
- Added voltage monitoring for HD 7950 and HD 7970
- Fixed memory size readings for ATI cards with large VRAM
- Improved formula for NVIDIA ASIC Quality reading
- Added explanation text to ASIC quality window
- Fixed bug that caused updater to show up even though no update available, lagging GPU-Z.
- When multi-GPU setup detected, PCIe load test will recommend full screen
- Added board ID to BIOS version readout
- Added option to show sensor reading in GPU-Z title (click the arrow next to the sensor name)
- Refresh sensors in background now defaults to enabled
- Fixed release date for HD 7950
- Added PCI vendor Packard Bell
- Fix for ATI hardware access breaking on Catalyst 12.1
- Added fan RPM monitoring support on some ATI cards
- Added GF108 based GT 520, GTX 555 (non-mobile), GeForce 305M, 610M
New in GPU-Z 0.5.8 (Jan 21, 2012)
- Added explanation about PCI-Express power savings and 3D render test to accurately measure bus config under load
- Added function to display ASIC quality for Fermi and Southern Islands. (Located in the GPU-Z system menu)
- Fixed crash on older ATI cards
- Added voltage monitoring for HD 7970
- Improved real-time clock monitoring for HD 7970
- Fixed OpenCL detection for AMD Antilles, Whistler, Seymour, Blackcomb
- Improved default clock reading for AMD HD 7970 and Fusion
- Added support for AMD FirePro V7900, HD 6930, HD 7690M, HD 6410D
- Fixed Intel Sandy Bridge IGP to be DirectX 10.1, 32 nm
- Added support for NVIDIA Tesla C2075, GeForce GT 630M
New in GPU-Z 0.5.7 (Dec 20, 2011)
- Added support for AMD Radeon HD 7970, HD 7350
- Fixed "APIC counter broken" message on AMD Fusion Graphics
- Fixed PCIe 3.0 showing up as 2.0
- Fixed fillrate calculation on Fermi architecture
- Fixed ROP count on GT 420, GT 520, HD 5450, HD 6450
- Fixed random values showing as default clocks on some NVIDIA cards
- Fixed random value showing as shader clock on NVIDIA cards without shader clock
- Added / corrected release dates for HD 3650, FirePro V4800 and V3800
- Added process size, die size, transistor count for Radeon E6760
- Added transistor count for FirePro V4900
- Improved detection for HD 5570, HD 6450A, HD 6470M
New in GPU-Z 0.5.6 (Nov 25, 2011)
- Removed PowerColor hardware giveaway
- Added support for ATI Radeon HD 6320, ATI FirePro V9800, AMD FirePro V4900
- Added support for GeForce GTX 560 Ti 448 cores, GeForce GTX 460 v2, NVIDIA Quadro NVS 420, NVIDIA Quadro NVS 450, NVIDIA Quadro FX 380 LP
- Fix for incorrect shader reading on Blackcomb
- Added basic detection for NVIDIA Kepler GK107
New in GPU-Z 0.5.5 (Sep 1, 2011)
- Added PowerColor hardware giveaway
- On multi-GPU systems, GPU-Z will remember the selected card from the selection dropdown
- Added full support for AMD A-Series Fusion processors
- Added detection for ASUS MARS II
- Fixed shader count detection for Blackcomb (mobile AMD Cayman)
- Improved BIOS version reading on NVIDIA
- Added support for NVIDIA GeForce GT 560 Ti OEM, GT 545, GT 530, GeForce GTX 580M, GT 555M, GeForce 520MX, GT 520M, 410M, 305M, Quadro 5000, Quadro 4000M, Quadro 400
- Added support for AMD HD 6290, Radeon E6760 & E4690, FirePro V8800, FirePro V3700, FirePro 2460 (FireMV)
New in GPU-Z 0.5.4 (Jun 29, 2011)
- Considerably improved startup time on ATI cards, faster on NVIDIA, too
- Improved rendering on high-DPI screens
- Fixed ROP count reading on AMD Radeon HD 6790, Turks & Whistler
- Fixed sensor count on Caicos, Whistler, Turks
- Fixed OpenCL detection on some NVIDIA drivers
- Memory size reading support for AMD Fusion GPUs
- Added die size and transistor count for GF108, GF114, Turks, Caicos
- Added detection for more Intel Sandy Bridge GPUs
- Added detection for variants of GT 220, GTS 450
- Added detection for faked cards based on GTS 250, 9600 GT, NV41M
- Added detection for GeForce GTX 560 non-Ti, GT 520, GF116 based GTS 450, GT 550M, GeForce 315, GeForce 405, Quadro 2000 & 6000
- Added detection for GeForce GTX 560M, GT 555M, GT 550M, GTX 540M, GT 525M, GT 520M, GTX 415M, GT 410M, Quadro 1000M, NVS 4200M
- Added detection for GT218 based ION 2 / 9300 GS / 8400 GS / 405 / NVS 300
- Added detection for HD 6750, HD 6770, HD 6850M, HD 6970M, HD 6670M, HD 6650M, HD 6330M, FirePro 2260
New in GPU-Z 0.5.3 (Apr 9, 2011)
- Added support for AMD Radeon HD 6790, FirePro V5800
- Added support for GeForce GT 555M, GTX 540M, G315M, Quadro FX 600, FX 2800M, Quadro 4000
- Fixed GeForce GTX 590 release date
- Added support for voltage controllers VT1556, uP6266, uP6218
- Added VDDCI measurement on CHL8228 based ATI cards
- Added capability to show sensors in TechPowerUp OSD Server
- Fixed several crashes and resource leaks
New in GPU-Z 0.5.2 (Mar 21, 2011)
- Added support for NVIDIA GT 440, GTX 550, GTX 560, GTX 570, GTX 590
- Added support for AMD HD 6990
- Added detection support for more Sandy Bridge IGPs
- Fixed ROP count on ATI Madison
- Improved OpenCL detection for ATI
- Added CHL8228 voltage controller support for Cayman Pro and XT
- Improved driver based voltage monitoring support for ATI cards
- Added startup parameter /card followed by the adapter index, starts with 1.
- Several crash and stability fixes
New in GPU-Z 0.5.1 (Jan 25, 2011)
- Added support for NVIDIA GTX 560 Ti, 480M, 485M
- Added transistors and die size for 460M, 445M, 435M
- Added release date for 6800 GT, 6600 GT, 6600, 6200, 8800 GTX, 8800 GTS, 8800 Ultra, 8600 GTS, 8500 GT, 8800 GS, 9600 GSO, 8800 GT, GTS 250, GT 440
- Added preliminary support for AMD Zacate and Ontario APUs
- Found a new way to disable NVIDIA's power limiter
- Added support for GTX 570 and GTX 560 to power limit disable function, the parameter is now called /GTX500OCP
- Fixed crash in sensors panel
- Acer PCI vendor now using a shorter text that fits the field width
- Added CHL8214 & CHL8226 support for AMD Barts and Cayman
- Improved the automatic update check to run completely in the background and only appear when an update is available (manual update check is unchanged)
New in GPU-Z 0.5.0 (Dec 27, 2010)
- Added full support for Radeon HD 6950 and HD 6970
- Fixed TMU/Texture Fillrate rounding error on NVIDIA cards
- Improved voltage monitoring support for GTX 570
- Workaround for ATI OpenCL driver crashing GPU-Z on systems with NVIDIA cards
- Fixed crash when driver version could not be read
- Fixed crash when splash screen interrupted by message box
New in GPU-Z 0.4.9 (Dec 3, 2010)
- Added support for GeForce GTX 570, GTX 460 SE
- Added support for PCI-Express 3.0
- Added PowerColor Giveaway
- Added Colorfire PCI vendor ID
- Added detection for AMD Radeon HD 6310, HD 6370
- Fixed BIOS >64k reading on GF110
- Added GT215 based GT 220
- Fixed TMU count on GF104
- Improved default clock reading on GeForce GTX 580
- Better error handling for memory type detection code
New in GPU-Z 0.4.8 (Nov 9, 2010)
- Added support for NVIDIA GeForce GTX 580 and future GF104 variants
- Updated TMU/Texture rate calculation method. Please report any errors in the forums
- Fixed memory clock monitoring on HD 6850, HD 6870
- Fixed label corruption on cards without any shaders
- Fixed rare CrossFire detection bug on ancient ATI drivers
- Added voltage monitoring fallback for ATI cards with unsupported voltage controller
- Updated PCI vendor ID for PowerColor
- Removed MSI giveaway
New in GPU-Z 0.4.7 (Oct 19, 2010)
- Improved support for AMD Radeon HD 6850 and HD 6870
- Added support for NVIDIA GeForce GT 430, GT 420
- Fixed default clock reading on Fermi architecture cards
- Added voltage controller support for Colorful iGame GTX 460
- Added PCI Vendor detection for Colorful
- Added TMU count and die size for GF106
- Synchronization mutexes are global now
- Improved stability when running multiple instances of GPU-Z / Fixed "the driver has been marked for deletion" error
New in GPU-Z 0.4.6 (Sep 2, 2010)
- Added MSI hardware giveaway
- Fixed temperature readings getting stuck on ADT7473 sensor (GTX 400 cards)
- Fixed broken BIOS reading on all cards
- General improvements to Intel support, still far from perfect
- Added support for ATI FirePro V8750
- Added support for GeForce GTS 455, GTS 450, GT 440, GT 430, GT 420, GT 330, GT218 based ION, GTX 470M, GTX 460M, GT 445M, GT 435M, GT 425M, GT 420M, GT 415M, GeForce 310M
- Added basic detection for AMD Radeon HD 6000 Series
- Improved handling for unknown NVIDIA cards
- Added fan RPM monitoring for NVIDIA cards
New in GPU-Z 0.4.5 (Aug 20, 2010)
- Fixed shader count on NVIDIA GTX 460, added die size & transistors
- Fixed ROP count on AMD Redwood
- Added support for ATI FirePro V7800, ATI IGP 320M, ATI FireGL X1, ATI HD 4200 variants
- Added launch on Windows startup function accessible via system menu and supporting UAC
- Autoupdater will not check for update when no Internet connection detected
- Fixed browser not starting correctly from autoupdater
- Screenshot detection that promotes GPU-Z's own screenshot upload feature
- Sensor data collection limited to 6 hours of data per sensor (~250 KB)
- Shortened Clevo PCI vendor name
New in GPU-Z 0.4.4 (Jul 1, 2010)
- Added support for NVIDIA GeForce GTX 460, GT 330, GT 315, GT216 based ION
- Added support for AMD HD 5670 (Juniper), Mobility HD 5430, Mobility 4200, Mobility 4100
- Added PowerColor hardware giveaway
- Added die size measurement for NVIDIA GF100
- Added option to GPU-Z's system menu to show temperatures in °F
- Fix for BIOS reading on ATI
- Fixed decimal points in ADT7473 reading
- Fixed crash on GeForce 8300 IGP
- Added clock monitoring for RS780 variants
- Improved clock monitoring for ATI R6xx architecture
New in GPU-Z 0.4.3 (May 21, 2010)
- Added full voltage controller monitoring for GeForce GTX 480
- Added real-time clock monitoring for GF100
- Added detection for vendor GALAXY
- Complete revamp of voltage controller architecture. Adds voltage monitoring for many ATI & NVIDIA cards and generic monitoring method for NVIDIA cards via driver.
- Added support for >4 GB memory on NVIDIA
- Fixed memory clock reading on some HD 5670 cards
- Fixed thermal sensor count on some ATI Evergreen models
- Fixed incorrect default clock reading on some ATI Evergreen models
- Added detection for NVIDIA GTX 465, GT 320M GT 335M, ION variant, GT 340, Tesla C2050, Tesla M2050, GTX 285M, GT 325M, Quadro FX 880M, Quadro FX 1800M
- Added detection for ATI HD 5570, Mobility 4550, Mobility 5870, HD 4250, HD 4290, FirePro V7750
- Added detection for Intel GMA 3150
- Added fix for ADT7473 coldbug (GTX 480) via commandline parameter /fixcoldbug
- Added OCP adjustment for CHL8266 (GTX 480). Use it via command line parameter /chl8266ocp without parameter to query, with numeric value seperated by space to set. This is a permanent change that will not be reverted by powering down the system.
- Added languages Hungarian and Spanish
- Updated languages French, Italian, Russian
New in GPU-Z 0.4.2 (Apr 7, 2010)
- Fixed an issue on all NVIDIA cards that would sometimes cause them to lose fan control when GPU-Z is running. Please don't run pre-0.4.2 and 0.4.2 at the same time, this might result in incorrect readings in the 0.4.2 window.
- Fixed system crash on MSI HD 5830 TwinFrozr II
- Validation upload now uses Unicode for entered name
- Reverted memory bandwidth calculation introduced in last version. It introduces too much confusion, we will go with what all companies use in their marketing material, even though we believe it is not correct. Using power of 10 again now, 1 GB/s = 1,000,000,000 Bytes per Second.
New in GPU-Z 0.4.0 (Mar 27, 2010)
- Added support for NVIDIA GeForce GTX 470 and GTX 480
- Added voltage monitoring for HD 5750, 5830, 5870 Lightning
- Added support for Intel Pineview
- Fixed double memory clock reading on NVIDIA w/ GDDR5 (real clock is now displayed)
- Fixed rounding errors in RV7xx and Evergreen fan monitoring code
- Added tooltip translations for Albanian and Chinese Traditional
- Memory bandwidth calculation uses GB now instead of GiB
- Added new hardware giveaway tab in cooperation with ASUS
New in GPU-Z 0.3.9 (Feb 23, 2010)
- Added localized tooltips in the languages: AM, AR, BG, CN, DE, FA, FR, IT, PT, RS, RU, TR, UA. Default language is your system language, you can change this setting in the system menu of GPU-Z.
- Added support for ATI HD 5830, HD 5770, HD 5750, HD 5670, HD 5570, HD 5450, HD 4860, HD 4750, Mobility HD 3430, M97, Mobile HD 5850, 5730, 5650, 5470 (Broadway, Madison and Park)
- Added support for NVIDIA GeForce 6200A, GTS 250, 9800S, 205, 310, GT 320, GT 240, GTS 360M, 230M, 330M, 240M, G105M, Quadro NVS170M, G210M, GTS 250M, future MCP, more ION variants.
- Added support for Intel GMA500, Clarkdale, Arrandale
- Added detection for variants of NVIDIA NV41, G94, G98, ATI RV350, RV370, R480, RV515. Most of them used on fake cards relabeled to NVIDIA 9x00 series.
- Improved default clock BIOS parsing on NVIDIA
- Fix for incorrect default clocks on HD 5000 Series
- Fixed incorrect real-time clock monitoring on some HD 5000 cards
- Added more accurate real-time clock monitoring function for NVIDIA
- Improved OpenCL detection on ATI R700, RV790, all HD 5000 cards
- Added temp. monitoring support for some Intel IGPs
- Improvements to CUDA detection
- Removed memory chip input fields from VGA BIOS Upload
New in GPU-Z 0.3.8 (Dec 1, 2009)
- Added framework for translations in GPU-Z, to do that we need your help. Please go to tpu.me/forums/forumdisplay.php?f=72 to submit contributions in your language
- Added sensors to monitor GPU load percentages on NVIDIA
- Fixed startup on Windows 2000 (DLL not found)
- Improved detection/added sensors for ATI M86
- Fixed several NVAPI crashes
- Fixed crash when PhysX not available
- GPU-Z can now be set in its system menu to be always on top of other windows
New in GPU-Z 0.3.7 (Nov 19, 2009)
- Added Infotips for sensors
- Fixed crash when ATI ULPS CrossFire enabled
- Added full support for ATI HD 5970 (Hemlock)
- Added full support for NVIDIA GeForce GT 240 (GT215)
- Added "ULPS active" note to CrossFire status when ULPS active
- Fixed overclocks via CCC not showing correctly
- Added checkboxes to show availability of CUDA, PhysX, OpenCL and DirectX Compute
- Improved memory type detection for GT 216, GT 218
- Added voltage monitoring support for HD 5770
- Disconnected fan now shows correctly in sensors (when detectable via hardware)
- Improved accuracy of ATI GPU load monitoring
- Added detection for ATI RV380/RV530 cards relabeled to NVIDIA 9400/9500/9600 GT
- Added support for NVIDIA 9300 SE, NVS450, GT 240, GTS 250M, G102M
- Added support for Intel G41
New in GPU-Z 0.3.6 (Oct 15, 2009)
- Added Infotips, they can be turned off in the system menu
- Added preliminary support for ATI Radeon HD 5900 (Hemlock), HD 5600 (Cedar), HD 5300 (Redwood)
- Added full support for HD 5850, HD 5770, HD 5750
- GPU-Z no longer crashes on ATI ULPS enabled CrossFire configurations
- Fillrate Pixel, Fillrate Texel, Memory Bandwidth will now correctly recalculate when clocks change
- Bus Interface, ATI shader count is now periodically refreshed
- Added support for HD 5850 voltages
- Second voltage on HD 5870 correctly named VDDCI
- ATI clock reading code fixed
- Fixed default clock reading on HD 5850
- Added Mutex to avoid collisions on long I2C operations
- Added basic detection for a lot of NVIDIA card fakes
- Added detection for ATI FireGL V7600, M92, RS880
- Added detection for NVIDIA GT 130, 9600 GE, Tesla C1060, G 220, GT 220, NVS 3100, GTS 160 M, NVS 160M, NVS 150M, G105M, GT 240M, G105M, G210M, G310M, ION LE
- Fixed die size and transistor count on G72 M
New in GPU-Z 0.3.5 (Sep 25, 2009)
- ATI Radeon HD 5870 (Cypress) fully supported (includes voltage monitoring)
- Added or improved support for NVIDIA FX5500, 9400 GT, G73, GTS 240, GT 140, FX 1800, GT 120 (Apple), FX 380, FX 350, GTX 295 Single PCB, Quadro CX, FX 5800, FX 4800, FX 3800, GTX 180M, GTX 260M, FX 2700M, G 110M, GT 120M, GT 220M, FX 1700M, G 105M, MCP79MX
- Added or improved support for ATI HD 4730, HD 4650 AGP, M92, M96, M98
- Added clock reading for Intel i910, i915, i945, 946
- Added support for DDR3 detection on G9x
- Added monitoring support for RV7xx based mobile chips
- Voltage controller "slaves" are now called "phases"
- Added Intel Havendale detection
- Fixed BIOS parsing on some newer HD 4870 cards
- Added voltage monitoring support for MSI N275GTX Lightning
New in GPU-Z 0.3.4 (Apr 29, 2009)
- Fixed RV740 Bugs: ROP count, shader count, real-time core clock, BIOS dump size
- Fixed number of transistors on RV790
- Fixed BIOS reading on RV790
- NVIDIA driver version is returned correctly on Windows 7
- GPU-Z should work again on Windows 2000 without PSAPI.DLL
- Added voltage reading support for HD 4890
- Fixed default clock reading on ATI cards that have default clock < boot clock
- Fixed GM45 to be 65 nm and DX 10.0
- Added release date for GTX 275
- Fixed GT 130M to be 55 nm and added its release date
New in GPU-Z 0.3.3 (Mar 25, 2009)
- Added support for RV740 (HD 4750, HD 4770) and RV790 (HD 4890)
- Fixed G98 shader count
- Fixed GT200 process size calculation
- Added support for GTX 275
New in GPU-Z 0.3.2 (Feb 18, 2009)
- Clock reading on ATI cards fixed
- GT200b is now detected as 55 nm
- Added detection for GeForce GT 120, GT 130, GT 130M, FX 5200, 9400 GT, 9600 GSO 512, FX 2700M, 9500 GS, Quadro FX 540M, Quadro FX 3700M, 9650M GT, Quadro FX 370M, 9400 IGP, 9300 IGP, 8200M G
- Added support for S3 Chrome 500 Series
- Added new ATI logo
- Fixed screen corruption during screenshot of sensors tab
- Added support to show lowest, highest and average reading on sensors tab (click on sensor value to switch)
- Fixed number of texture units on RV730
- Fixed Catalyst version reading
New in GPU-Z 0.3.1 (Jan 5, 2009)
- Fixed crash on systems without NVAPI (Win 2K, old NV drivers)
- Optimized RV770 BIOS reading code for speed (faster application startup)
- Improved voltage reading code for recent ATI cards
- Added support for fan speed monitoring on ForceWare 180.xx and up
- Added voltage monitoring for NVIDIA (GT200)
- Changed reading method for RV770 sensors, fixes race conditions with other software and on-board fan control
- NVIDIA 9800 GTX+ listed as 55nm now
- Added support for Intel Q43/Q45
- Added support for NVIDIA Quadro FX 4700 X2, GTX 295, GTX 285
- Added preliminary support for NVIDIA GT212, GT214, GT215, GT216, GT218
New in GPU-Z 0.3.0 (Dec 4, 2008)
- Fixed monitors panel showing up sometimes Re-enabled dupecheck for VGA BIOS upload Corrected 9800 GTX release date Added support for NVIDIA Quadro VX 200 Fixed voltage readings on HD 4870 and HD 4870 X2
New in GPU-Z 0.2.9 (Dec 4, 2008)
- Added support for NVIDIA 9800 GT, 9600 GS/GSO, 9300 GS, Quadro FX Go1400, 9800M GTX, 9600M GT, 9650M GT, Quadro FX 770M, 9200M GS, 9300M GS, Quadro NVS 160M, G71 0x2e5
- Added support for ATI HD 4350, HD 4550, HD 4650, HD 4830, HD 4850 X2, RV670 0x9504
- Added support for Intel GM45 variant 0x2a43
- Added support for shader count reading on HD 48xx
- Changed ATI driver function call method. This should result in more accurate readings on X2 cards and non-desktop ATI devices
- Improved VGA BIOS Upload dialog. Drop down lists appropriate for your card will now be fetched from our servers to facilitate the upload process
- Corrected RV610 transistor count
New in GPU-Z 0.2.8 (Sep 24, 2008)
- Added logging of sensor data to file
- Added option to disable polling for individual sensors
- Added support for AMD RV730
- Added clock reading support for Catalyst 8.9 on RV770 and RV730
- Added support for NVIDIA 9400 GT, 9500 GT, 9600 GS, 9800M GS, 9600M GS, 9300M GS, 9200M GE
- Added support for AMD R580 Stream Processor, RV730 and one M88 variant
- Fixed PCI subvendor ID 1787 to be ATI AIB
- Fixed readings and added monitoring support for ATI M88
- Fixed fan speed readings on ADT7473 sensor
- Added support for ADT7473 sensor chip on ATI
- Added release dates for several NVIDIA mobile GPUs
New in GPU-Z 0.2.7 (Aug 13, 2008)
- Added display for GPU release date. The data is incomplete, if you have more info please post in the sticky thread in the support forums.
- Added display for transistor count
- Added mouseover readings for sensors graph
- Correctly named the three on-die sensors of RV770 and RV730
- Added support for voltage regulator monitoring on HD 4870 and HD 4870 X2
- RV770 GPU on HD 4870 X2 now called R700
- Added support for reading HD 4870 X2 3D clocks
- Added display of CrossFire Sideport status on HD 4870 X2
- Added more RV730 device IDs
- Added support for GeForce 9800 GT, 9800M GT, 9600M GT, 9700M GS/GT, 9500M G
- Added detection for RS780D
- Fixed incorrectly read ATI OD5 clocks
- Fix for memory clock reading on RV730
- Fixed ROM reading on RV730
- Fixed crash when no supported card detected
- Sensor graph width adjusted when scrollbars visible
New in GPU-Z 0.2.3 (Jun 1, 2008)
- Fixed NVIDIA GeForce GTX 260/280 (GT200) support. Everything is working and tested now
- Removed bogus 190�C PCB temp sensor on various NVIDIA cards
- Added fan speed monitoring for NVIDIA
- Fixed crash when switching devices during sensor refresh
- Fixed ROP/Shader count on NVIDIA G72
- Fixed graphs spilling over their client area
- Fixed a number of other rare crashes
- NVIDIA device ID 0x00F3 now correctly recognized as NV43
- Improved crash reporting module
New in GPU-Z 0.0.4 Beta (Oct 6, 2007)
- Fixed 8800 GTS, 8500, 8600 misdetection