Does Stress Cache only stress the Cache ?
And if there's an error, is it necessarily due to Vcache, or could it come from something else ?

Thanks

For the drive in my screenshot below, the average is shown as 1598.3MB/s -- which seems right considering the shape of the graph and my experience with the drive.  However, I was trying to validate the Average speed reported by the tool and came up with some inconsistencies. 

I'm operating under three main assumptions:

1. The benchmark writes the full capacity of the drive, no more, no less.
2. There isn't any idle time between block writes or elsewhere in the test.
3. The timer is accurately reporting the length of the benchmark

If you take the number of MB written -- the size of the drive -- in this case 976,793.6MB (953.9GB * 1024) and divide that by the time elapsed 1031sec ((17min * 60) + 11) you get 947MB/s.

I performed another iteration of this test and used a camera to record the whole benchmark so I could validate how long it took (I didn't trust myself with a stopwatch), and the time reported by the benchmark ~17min was pretty close to the amount of time measured in my recording... so the issue isn't related to the timer in the benchmark reporting the wrong time.

In Windows I've seen this drive definitely be able to sustain writes over 1.3GB/s (anecdotally), so my calculated 947MB/s definitely doesn't seem right, but then that means that one of my first two assumptions is wrong -- either the timer in the benchmark is including idle time, extending how long it takes to perform the disk write, or the test is performing more writes than the full capacity of the drive.  The latter of these two seems unlikely, but that's just a gut feeling.

--

Can you provide any insight into how the tool calculates the average speed or any thoughts as to why this discrepancy exists?  Is there a way to export the datapoints used for the calculations?

Thanks!!

I've exhausted all Aida64 benchmarks (memory + cpu).

My question is :  when testing CPU Queen perfs for example, the cpu frenquency displayed is 3900 Mhz with overclock due to turbo-boost (i7-3770 not K, base is 3.4 Ghz). All other cpus display results and base frequencies associated, let's say 3500 Mhz for i7-3770K (base 3.5 Ghz, T.B. 2.0 is 3.9 Ghz).

Does this means that all other cpu compared are executed with turbo-boost disabled ? or only real frequency is displayed for the cpu which is tested ? Or, in another words, do I have to disable turbo-boost to benchmark my cpu ?

Thank's for clarify those infos...

Memory benchmarks show incorrect frequency (ddr5-3197) and timings.

In fact, ddr5-5600 memory modules are installed.

I assume that Aida shows the frequencies and timings in idle time, instead of the values during the benchmark.

When I check the box "Stress GPU (s), I have this popup (see screen)
what is it exactly ?

Thanks

Which Nvidia driver file (or dll) I guess is necessary for this test ?
I don't know since when I've had this "error," before it was fine.

Thanks

But most importantly, we're introducing a brand new benchmark panel that offers a set of OpenCL GPGPU benchmarks that you can launch from AIDA64 / main menu / Tools / GPGPU Benchmarks.  These benchmarks are designed to measure GPGPU computing performance via different OpenCL workloads.  Every benchmark methods are designed to work on up to 16 GPUs, including AMD, Intel and nVIDIA GPUs, in any combination.  Of course CrossFire and SLI configurations, and both dGPUs and APUs are also fully supported.  HSA configurations are handled via preliminary support.  Basically any computing capable device will be benchmarked that appears as a GPU device among OpenCL devices.

The OpenCL benchmark methods currently offered are not specifically optimized for any GPU architectures. Instead, the AIDA64 OpenCL module relies on the OpenCL compiler to optimize the OpenCL kernel to run best on the underlying hardware.  The OpenCL kernels used for these benchmarks are compiled in real-time, using the actual OpenCL driver the OpenCL GPU device belongs to.  Due to that approach, it is always best to have all video drivers (Catalyst, ForceWare, HD Graphics, etc) updated to their latest & greatest version.  For compilation the following OpenCL compiler options are passed: -cl-fast-relaxed-math -cl-mad-enable.

On top of that, the GPGPU Benchmark Panel also has a CPU column, for comparison purposes.  The CPU measurements however are not obtained via OpenCL, but using native x86/x64 machine code, utilizing available instruction set extensions like SSE, AVX, AVX2, FMA and XOP.  The CPU benchmarks are very similar to the old CPU and FPU benchmarks AIDA64 has got, but this time they measure maximum computing rates (FLOPS, IOPS).  The CPU benchmarks are heavily multi-threaded, and are optimized for every CPU architectures introduced since the first Pentium came out.

The following benchmark methods are currently offered.  We've indicated the x86/x64 CPU benchmark difference in brackets where there is a different approach in benchmarking.

1) Memory Read: Measures the bandwidth between the GPU device and the CPU, effectively measuring the performance the GPU could copy data from its own device memory into the system memory.  It is also called Device-to-Host Bandwidth.  [[[ The CPU benchmark measures the classic memory read bandwidth, the performance the CPU could read data from the system memory. ]]]

2) Memory Write: Measures the bandwidth between the CPU and the GPU device, effectively measuring the performance the GPU could copy data from the system memory into its own device memory.  It is also called Host-to-Device Bandwidth.  [[[ The CPU benchmark measures the classic memory write bandwidth, the performance the CPU could write data into the system memory. ]]]

3) Memory Copy: Measures the performance of the GPU's own device memory, effectively measuring the performance the GPU could copy data from its own device memory to another place in the same device memory.  It is also called Device-to-Device Bandwidth.  [[[ The CPU benchmark measures the classic memory copy bandwidth, the performance the CPU could move data in the system memory from one place to another. ]]]

4) Single-Precision FLOPS: Measures the classic MAD (Multiply-Addition) performance of the GPU, otherwise known as FLOPS (Floating-Point Operations Per Second), with single-precision (32-bit, "float") floating-point data.

5) Double-Precision FLOPS: Measures the classic MAD (Multiply-Addition) performance of the GPU, otherwise known as FLOPS (Floating-Point Operations Per Second), with double-precision (64-bit, "double") floating-point data.  Not all GPUs support double-precision floating-point operations.  For example, all current Intel desktop and mobile graphics devices only support single-precision floating-point operations.

6) 24-bit Integer IOPS: Measures the classic MAD (Multiply-Addition) performance of the GPU, otherwise known as IOPS (Integer Operations Per Second), with 24-bit integer ("int24") data.  This special data type are defined in OpenCL on the basis that many GPUs are capable of executing int24 operations via their floating-point units, effectively increasing the integer performance by a factor of 3 to 5, as compared to using 32-bit integer operations.

7) 32-bit Integer IOPS: Measures the classic MAD (Multiply-Addition) performance of the GPU, otherwise known as IOPS (Integer Operations Per Second), with 32-bit integer ("int") data.

8) 64-bit Integer IOPS: Measures the classic MAD (Multiply-Addition) performance of the GPU, otherwise known as IOPS (Integer Operations Per Second), with 64-bit integer ("long") data.  Most GPUs do not have dedicated execution resources for 64-bit integer operations, so instead they emulate the 64-bit integer operations via existing 32-bit integer execution units.  In such case 64-bit integer performance could be very low.

9) Single-Precision Julia: Measures the single-precision (32-bit, "float") floating-point performance through the computation of several frames of the popular "Julia" fractal.

10) Double-Precision Mandel: Measures the double-precision (64-bit, "double") floating-point performance through the computation of several frames of the popular "Mandelbrot" fractal.  Not all GPUs support double-precision floating-point operations.  For example, all current Intel desktop and mobile graphics devices only support single-precision floating-point operations.

------------------------------------------------------------------------

As for the GPGPU Benchmark Panel's user interface:

1) You can use the checkboxes to enable or disable utilizing a specific GPU device or the CPU.  The state of the CPU checkbox is remembered after closing and re-opening the panel.

2) You can launch the benchmarks for the selected devices by pushing the Start Benchmark button.  In case you want to run all benchmarks, but only on the GPU(s), you can double-click on the GPU column label to do so.  In case you only want to run the Memory Read benchmarks on both the GPU(s) and the CPU, you can double-click on the Memory Read label to do so.  In case you only want to run the Memory Read benchmark on only the GPU(s), you can double-click on the cell where the requested result should appear after the benchmark is completed.

3) The benchmarks are executed simultaneously on all selected GPUs, using multiple threads and multiple OpenCL context, each with a single command queue.  CPU benchmarks however are only launched after the GPU benchmarks are completed.  It is currently not possible to run the GPU and CPU benchmarks simultaneously.

4) In case the system has multiple GPUs, the first results column will display an aggregated score for all GPUs.  The individual GPU results are combined (added up), and the column label will read e.g. "4 GPUs".  If you want to check the individual results, you can either uncheck some of the GPUs until just one GPU is left checked, or push the Results button to open the results window.

5) In case you've got exactly two GPU devices, and you disable the CPU test by unclicking its checkbox, the panel will switch to dual-GPU mode where the first column will be used for GPU1 results, and the second column will be used for GPU2 results.  If after obtaining the results you want to check the combined performance of GPU1+GPU2, just check the CPU again, and the interface will switch back to the default layout.

------------------------------------------------------------------------

FAQ:

Q: Is it possible to measure performance of OpenCL CPU devices?

A: No, it's not available currently, because OpenCL CPU drivers are simply not suitable for proper benchmarking.  They execute code a lot slower than native x86/x64 machine code or sometimes even regular multi-threaded C++ code.

Q: Do AIDA64 GPGPU benchmarks use vectorized data types and unrolling techniques to boost performance?

A: Yes, both, in order to make the job of OpenCL compilers a bit easier.  On top of that, the OpenCL compiler may still use additional optimizations, like further unrolling, it is completely up to the OpenCL compiler.

Q: Is the OpenCL-capable VIA chipset (VX11) supported?

A: No, because currently there's no stable OpenCL compiler and OpenCL driver for VIA chipsets or processors.

Q: Are OpenCL 2.0 and HSA supported on AMD Kaveri systems?

A: Yes, except for the memory benchmarks.  Memory benchmarks currently don't work with HSA, because the current AMD HSA implementation doesn't yet support forcing the usage of device memory, but instead it automatically assumes that allocated memory blocks are to be shared between the CPU and GPU.  As soon as AMD's OpenCL 2.0 and HSA implementation gets more mature, these issues will be resolved.

Q: Are the latest generation dGPUs, like AMD Radeon R9 290/290X, nVIDIA GeForce GTX Titan and GTX 780 fully supported?

A: Yes, but on such dGPUs where clock boosting and/or throttling is used, it is very important to decide whether you want to measure the absolute maximum attainable performance, or the average performance.  If you're looking for the absolute maximum scores, then make sure to start AIDA64 GPGPU Benchmarks when the video card is cool, and with power limits set to a relaxed value (AMD PowerControl).  If you're looking for the average performance, then make sure to disable the CPU benchmarks, and execute the GPU benchmark methods at least 10 times right after each other, to properly heat the video card up.

Q: Is OpenCL benchmarking under Windows 8.1 and Windows Server 2012 R2 supported?

A: Yes, as long as the video drivers are properly installed.

Q: On the Intel Core i7 "Haswell" processor, the CPU results are all considerably higher than the Intel HD Graphics 4600 "GT2" GPU results.  How is that possible?

A: AIDA64 CPU benchmarks are heavily optimized for Haswell and all other modern CPU architectures, and they utilize any available instruction set extensions like SSE, AVX, AVX2, FMA or XOP, and of course full vectorization as well.  Using FMA and AVX2, a quad-core Haswell's x86/x64 part can indeed provide very high computing performance, well exceeding the performance of its GT2 iGPU.  It is however much easier to write such optimized code for the iGPU via OpenCL, than for the CPU via machine code generator or x86/x64 assembly.

------------------------------------------------------------------------

You can try the new OpenCL GPGPU Benchmarks in the following new beta release of AIDA64 Extreme:

http://www.aida64.com/downloads/aida64extremebuild2656b7hl0kzgtszip
 

After upgrading to this new version, make sure to restart Windows to finalize the upgrade.

Please let us know here in this topic if you've got any comments or ideas about the new benchmarks.

AIDA64 doesn't seem to work well with my nVIDIA RTX5080 Founders Edition.

When I run the system stability test, the 5080 is not tested. There is no change to clock/mem/utilization, no change in fan speed or temperature.

If I have onboard iGPU enabled in my UEFI, then stability test will check the onboard Radeon GPU only.

With iGPU disabled in UEFI, the stability test does not appear to do anything.
However, during the Stability Test, the sensors for the nVIDIA RTX5080 *are *displayed. Just that the card isn't assigned any load during the stability test. How can I run a successful stability or GPGPU test on my 5080?  Honestly this is the main reason I purchased Extreme.

In the GPGPU Benchmark, the GPU is Not Supported (see attached screenshot) at all.

• AIDA64 Extreme
• Product key: 3XJPD-MFCD6-JVN2Z-EDPY1-JAC5T
• OS: Windows 11

• ------[ Versions ]------

  Program Version : AIDA64 Extreme v7.60.7317 Beta
  LLKD Version    : 1014
  BenchDLL Version: 4.7.914.8-x64
  Windows Version : Microsoft Windows 11 Pro 10.0.26100.3476 (64-bit)
  GetProductInfo  : 0x00000030

  ------[ Video Adapters ]------

  nVIDIA GeForce RTX 5080 Founders Edition [10DE-2C02 / 10DE-2095 / Rev A1]

  ------[ Video Driver ]------

  atiadlxx: 32.0.13031.3015 - AMD Adrenalin 25.3.1 WHQL

  nvapi.dll: 32.0.15.7283 - nVIDIA ForceWare 572.83

  C:\WINDOWS\System32\DriverStore\FileRepository\nv_dispi.inf_amd64_9d15b9aa9e1c885b\nvldumdx.dll|||C:\WINDOWS\System32\DriverStore\FileRepository\nv_dispi.inf_amd64_9d15b9aa9e1c885b\nvldumdx.dll|||C:\WINDOWS\System32\DriverStore\FileRepository\nv_dispi.inf_amd64_9d15b9aa9e1c885b\nvldumdx.dll|||C:\WINDOWS\System32\DriverStore\FileRepository\nv_dispi.inf_amd64_9d15b9aa9e1c885b\nvldumdx.dll|||C:\WINDOWS\System32\DriverStore\FileRepository\nv_dispi.inf_amd64_9d15b9aa9e1c885b\nvldumd.dll|||C:\WINDOWS\System32\DriverStore\FileRepository\nv_dispi.inf_amd64_9d15b9aa9e1c885b\nvldumd.dll|||C:\WINDOWS\System32\DriverStore\FileRepository\nv_dispi.inf_amd64_9d15b9aa9e1c885b\nvldumd.dll|||C:\WINDOWS\System32\DriverStore\FileRepository\nv_dispi.inf_amd64_9d15b9aa9e1c885b\nvldumd.dll

  ------[ nVIDIA GPU #1 @ mem D8000000 ]------

  Video Adapter = nVIDIA GeForce RTX 5080 Founders Edition
  GPU Code Name = GB203-400-A1
  PCI Device    = 10DE-2C02 / 10DE-2095 [Rev A1]

  GPU Clocks   = 2797 / 0 / 0 MHz
  Memory Clock = 15184 MHz

  GPU    Base Clock = 13500 kHz
  Shader Base Clock = 13500 kHz
  Memory Base Clock = 13500 kHz

  nvidia-000000 = 1B3000A1  [GB203]
  nvidia-088000 = BADF1100  []
  nvidia-08A000 = FFFFFF8F

  ------[ nVIDIA GPU #1 / ForceWare Properties ]------

  Memory Type             = 16 (GDDR7)
  Memory Vendor           = 1 (Samsung)
  Memory Bank Count       = 1
  Memory Bus Width        = 256-bit
  GPU / MC / VE / BI Load = 3 / 2 / 0 / 18
  VPE Count               = 0
  Shader Pipe Count       = 7
  Shader SubPipe Count    = 42
  Total SP Count          = <unknown>
  Total TPC Count         = 0
  Total SM Count          = <unknown>
  Core Count              = 10752
  ROP Count               = 112
  Partition Count         = 4 (mask: 0000000F)
  Fan RPM                 = <unknown>
  Fan Duty Cycle          = <unknown>
  All Fans RPM & Duty     = 1200 RPM / 1198 RPM / 30 % / 30 %
  Core Voltage            = <unknown>
  Core Rail Voltage       = 860000 uV
  Current Pstate          = 0
  PerfCap Reason          = 4 (Reliability Voltage)
  Ventura State           = <unknown>

  Physical Frame Buffer Size = 16693824 KB
  Virtual  Frame Buffer Size = 16694272 KB
  Total / Free Video Memory  = 16694272 KB / 14310616 KB

  ------[ nVIDIA GPU #1 / ForceWare Clocks (old) ]------

  Perf. Levels   = 5
  Active Profile = 4

  curr-core/shader/mem-0 = 937 MHz / 0 MHz / 405 MHz
  curr-core/shader/mem-1 = 937 MHz / 0 MHz / 810 MHz
  curr-core/shader/mem-2 = 937 MHz / 0 MHz / 7001 MHz
  curr-core/shader/mem-3 = 937 MHz / 0 MHz / 14984 MHz
  curr-core/shader/mem-4 = 937 MHz / 0 MHz / 15184 MHz [Active]

  orig-core/shader/mem-0 = 180 MHz / 0 MHz / 405 MHz
  orig-core/shader/mem-1 = 180 MHz / 0 MHz / 810 MHz
  orig-core/shader/mem-2 = 180 MHz / 0 MHz / 7001 MHz
  orig-core/shader/mem-3 = 180 MHz / 0 MHz / 14801 MHz
  orig-core/shader/mem-4 = 180 MHz / 0 MHz / 15001 MHz [Active]

  ------[ nVIDIA GPU #1 / ForceWare Clocks (old) ]------

  Perf. Levels   = 5
  Active Profile = 4

  curr-core/shader/mem-0 = 937 MHz / 0 MHz / 405 MHz
  curr-core/shader/mem-1 = 937 MHz / 0 MHz / 810 MHz
  curr-core/shader/mem-2 = 937 MHz / 0 MHz / 7001 MHz
  curr-core/shader/mem-3 = 937 MHz / 0 MHz / 14984 MHz
  curr-core/shader/mem-4 = 937 MHz / 0 MHz / 15184 MHz [Active]

  orig-core/shader/mem-0 = 180 MHz / 0 MHz / 405 MHz
  orig-core/shader/mem-1 = 180 MHz / 0 MHz / 810 MHz
  orig-core/shader/mem-2 = 180 MHz / 0 MHz / 7001 MHz
  orig-core/shader/mem-3 = 180 MHz / 0 MHz / 14801 MHz
  orig-core/shader/mem-4 = 180 MHz / 0 MHz / 15001 MHz [Active]

  ------[ nVIDIA GPU #1 / ForceWare Clocks (new) ]------

  Perf. Levels   = 1
  Active Profile = 255

  curr-core/mem-0 = 2707 MHz / 15184 MHz

  orig-core/boost/mem-0 = 2295 MHz / 2617 MHz / 15001 MHz

  ------[ nVIDIA GPU #1 / ForceWare Clocks (new) ]------

  Perf. Levels   = 1
  Active Profile = 255

  curr-core/mem-0 = 2775 MHz / 15184 MHz

  orig-core/boost/mem-0 = 2295 MHz / 2617 MHz / 15001 MHz

  ------[ NVAPI GPU Handles ]------

  GPU Count = 1

  GPU #1 = 00000100

  ------[ nVIDIA GPU #1 / Thermal Settings V2 ]------

  Sensor Count = 1

  Sensor #0 / Controller   = GPU Internal
  Sensor #0 / Target       = GPU Core
  Sensor #0 / Current Temp = 31 Celsius

  ------[ nVIDIA GPU #1 / Thermal Settings V1 ]------

  Sensor Count = 1

  Sensor #0 / Controller   = GPU Internal
  Sensor #0 / Target       = GPU Core
  Sensor #0 / Current Temp = 31 Celsius
   

Thank you very much.

https://rog-forum.asus.com/t5/intel-800-series/extreme-and-memory-latency/td-p/1095362

But more importantly, we've replaced the outdated set of memory benchmarks with brand new ones.  The new bandwidth (read, write copy) benchmarks now use multiple threads to squeeze out every last bit of performance from the caches and the memory modules.  On modern multi-core processors, using the old single-threaded benchmarks you couldn't see the actual memory bandwidth, but only the memory bandwidth available for single-threaded applications.  With the new benchmarks you will however get considerably higher scores, much closer to the theoretical memory bandwidth available.  It is especially true for 3-channel and 4-channel memory configurations, such as Intel X58 and X79 based high-end desktop systems; and also for NUMA-enabled multi-socket systems, such as 2- and 4-way AMD Opteron and Intel Xeon based servers and workstations.  For example:

Core i7-3960X with X79 chipset and 4-channel DDR3-1600:

- AIDA64 v2.85 Memory Read: 16825 MB/s  [ old ]

- AIDA64 v3.00 Memory Read: 45640 MB/s  [ new ]

We've also implemented multi-threaded cache benchmarks, that now include support for the latest AVX and AVX2 instruction set extensions.  Thanks to using all CPU cores for the cache benchmarks, you will get dramatically different cache bandwidth scores than with the old benchmarks.  For example:

Core i7-3960X with X79 chipset and 4-channel DDR3-1600:

- AIDA64 v2.85 L1 Cache Read: 121.8 GB/s  [ old ]

- AIDA64 v3.00 L1 Cache Read: 674.7 GB/s  [ new ]

And thanks to the 2x widened L1 cache bandwidth of Intel Haswell, on these new processors using the new cache benchmarks of AIDA64 v3.00 you will get unusually high scores.  Using Haswell, with a hint of overclock it's quite easy to cross the 1 TB/s mark for the L1 cache

We've also replaced the old cache and memory latency benchmark with a brand new one that uses a different approach, recommended by processor architecture engineers.  The old memory latency benchmark used the classic forward-linear solution, so it "walked" the memory continuously, in forward direction.  Unfortunately that classic approach was sometimes over-optimized by "too smart" memory controllers, that led to unrealistically low latency scores.  It was a constant fight for us to get around those over-optimizations, to make sure AIDA64 provides stable and reliable latency results.  With the new latency benchmark we've switched to a block-random solution, that keeps "jumping" to random addresses inside a memory block for a period of time, and then skips to a new block and continues "jumping" to random places inside there as well.  With this new solution memory controllers cannot find a pattern anymore in the latency measurement, and so they cannot over-optimize the benchmark.  The block-random approach however means that latency results will be higher, and since the scores are in nanosec, it means the results will be worse than what you got used to.  For example:

Core i7-3960X with X79 chipset and 4-channel DDR3-1600:

- AIDA64 v2.85 Memory Latency: 55.9 ns  [ old ]

- AIDA64 v3.00 Memory Latency: 67.5 ns  [ new ]

AIDA64 v3.00 also supports benchmarking the eDRAM L4 cache of the upcoming Intel Crystal Well processors.  An interesting article about Crystal Well:

http://www.anandtech.com/show/6993/intel-iris-pro-5200-graphics-review-core-i74950hq-tested

Less radical change, but still deserves a note: we've also revamped the framework around the CPU ZLib benchmark, so you can see a few percent gain in performance using AIDA64 v3.00.  For example:

Core i7-3960X with X79 chipset and 4-channel DDR3-1600:

- AIDA64 v2.85 CPU ZLib: 418.3 MB/s  [ old ]

- AIDA64 v3.00 CPU ZLib: 444.5 MB/s  [ new ]

We hope you'll find the new AIDA64 release useful.  Feel free to ask any technical or not-so-technical questions about AIDA64 v3.00 benchmarks in this topic.

Regards,

Fiery

 

Update:

Logs posted below

---------

On my 9700X + Gigabyte Aorus Elite, 16Gx2 6000 C28 system ,

Win11, 10.0.26100.3476

VBS off

AIDA64 7.60.7300 got that incredible result, as well as CPU latency increases

By https://github.com/ChipsandCheese/MemoryLatencyTest/releases , the second pic

By intel MLC, the 3rd pic.

If I turned VBS on, AIDA64 back to a normal result.

I don't know which is correct.

I'm a little worried after trying to test my system last night.

It's a new machine Ryzen 9900x, Aorus x870 elite motherboard, rx 9070 Steel Legend OC GPU, G.SKill Trident Z5 Neo RGB 32GB (2 x 16GB) DDR5 6000MHz CL30 Desktop RAM, Thermalright Hypervision AIO. 

With every other stability test I've run it's passed with no problems, Prime95, OCCT, Cinebench24 etc., temp never goes over 80c and averages around 65-70c on full load for each program.

When I try to run Aida64 temps spike to 90-92c and I get severe system lag (Stress CPU, FPU, cache, system memory). I can't move my mouse without significant lag. Eventually I can stop test. I only have EXPO profile enabled in my BIOS.

This is the first time I've used Aida64 so it might be user error, but I thought I'd reach out regardless.

Cheers

I built a new system with a 9800X3D, ASRock x870E Nova, and 32GB of DDR5 6000 CL30 a few days ago and have been testing for system stability and a potential undervolt. I noticed that my initial undervolt settings (-30 CO, +200 Mhz) seemed stable in a few benchmarks but then threw the hardware failure error on the default settings Aida64 stress test within a few seconds of starting. I then set disabled PBO, disabled EXPO, and turned off any other OC-related settings in the BIOS to check stock system stability. To my surprise, they system still fails the Aida64 stress test within 5 seconds of starting even on stock CPU settings without any OC/UV.

I also tried only stress testing CPU/FPU or CPU/FPU/cache, and at least twice this has caused the system to freeze up within a few seconds of starting the test, requiring a manual restart via the power/reset button. If i just select one test at a time (e.g. CPU only, FPU only, cache only) then the tests seem to run without issues. The system appears otherwise stable in benchmarks and stress tests such as OCCT, Cinebench R23, CPU-Z, so I'm trying to figure out if this is a bug with Aida64 or an actual hardware issue that may require an RMA. For reference, I am using Aida64 7.60.7300.

Any suggestions or feedback would be appreciated!

"FPU VP8" benchmark is very strange and do not scale on multi-core systems.

May be replace it to a more common HEVC encoding?

1) CPU Queen

Why is the result of the newest Threadripper 7975WX worse than that 3970X (and even old 2990WX)?

2) FPU Mandel

Why is the result of the newest Threadripper 7975WX worse than that 3970X? it should be twice as fast, as can be seen from the result in FP64 Raytrace.

Here's specs:

Intel Core i9-13900K
ASUS ROG Maximus Z790 HERO
NVIDIA RTX 3080 Ti Founders Edition
Kingston Renegade RGB DDR5-6400
Boot drive: Samsung 980 PRO 256GB
Tested drive: (in this screenshot)
OS: Windows 11 Home 22H2

https://imgur.com/a/norBJnI

Thinking it might be a one-time issue, I tested my old 16GB kit (rated at 3200MHz), and to my surprise, it also only runs at 2400MHz.

I've tried everything I could think of in the BIOS – enabling XMP/DOCP, adjusting settings manually, and double-checking for any overlooked options – but nothing seems to work.

I’m out of ideas now and would appreciate any suggestions or insights on what might be causing this and how to get the RAM running at its correct speed.

System Specs:

CPU: Ryzen 5 5600x

Motherboard: B450M/ac R2.0

RAM: 32GB 3600MHz (New), 16GB 3200MHz (Old)

GPU: RX 7700 XT

PSU: 550W

We have been using Aida64 report tool for years with great outcome. There is one item in the report that we cannot find or know how to generate a report of.

It is the Extended version of the Windows Services (services.msc) application. 

Could you please guide me through on how to get this report?

Kind Regards,

Guillermo Cisneros

do you plan to add AI/ML benchmark for CPU and GPU (using Intel AMX and Nvidia Tensor Cores, also Ryzen AI and Intel VPU)?

While all other readings seem fine, the L3 cache scores are far lower than other users of the same processor. Some with quite a big difference, for instance a few images i saw had 500+ GB/s read speed.

Is there something i'm doing wrong? Maybe there's a Motherboard bios setting related to this? Can this be a faulty CPU? Temps are also very low so it's not overheating.