Jump to content
HWBOT Community Forums

_mat_

Members
  • Content Count

    552
  • Joined

  • Last visited

  • Days Won

    2

Everything posted by _mat_

  1. _mat_

    Tweak book for cheats

    The crash during the online submission in GPUPI happens because of this problem: It should be possible to fix it on the bot side of the interface. In any case I have already improved the online submission in GPUPI 4 to avoid crashes like these.
  2. I don't know when it stopped working, but I am getting reports since January that GPUPI crashes when submitting online while data files are working. As websmile was able to submit a result with y-cruncher half an hour ago, I guess the problems are related to a specific combination of OS/hardware. I am currently working with Win10 RS5 + 9900K/Z390 and 2990WX/X399. Both platforms have the same submission problems. Do you have a detailed error log for these HTTP 400 errors? The response body is an HTML error page btw but no specific information on it. I am guessing it is triggered by the web application.
  3. The submission API currently returns a 400 error. No specific error message. Doesn't work with GPUPI, y-cruncher and x265. @richba5tard @Leeghoofd
  4. Pushing this. Online submission is not working, guys! @richba5tard
  5. In the last few weeks I've been working on a benchmark to calculate pi on graphics cards. It's an idea that's in my head for some years now, but I couldn't put it to life until now. Somehow the death of our beloved Turrican inspired me - I wanted to do something good, something that matters. So I tried hard and it finally worked out. I want to dedicate this benchmark to Turrican, like a virtual monument to never forget his legend. A benchmark in his honour. Together with the community of overclockers.at I've tested the benchmark for the last week and put out some new versions, wrote a technical description and added some frequently asked questions. It's still a beta version, but it should be pretty stable with most of the graphics cards and processors around. But be sure to install the latest GPU/OpenCL drivers for the target device, old ones can be very bugged. Here are a few examples for calculating the billionth digit of pi: GPUPI 1B: AMD Radeon R9 290, NVIDIA GeForce GTX 980 and Intel Core i7-4960X@4 GHz I think that there's a lot of potential in a benchmark like this, it's just fun to crunch pi on a gpu. If you like it, I would consider implementing multi gpu support and hwbot direct submission. Bugs and feature suggestions can be made here or in the international support thread at our forums. Download, technical details & FAQ: GPUPI
  6. _mat_

    Tweak book for cheats

    GPUPI 4 will be equally fast on GPUs, but faster on CPU when using the native path without OpenCL. That's my goal and it's necessary to leave OpenCL behind once and for all. The 3.3 categories should then be renamed to 4.0 (or 3.3+) to host all new GPUPI versions. The whole situation kind of sucks balls, I should have never released 3.3 so early. Would have been better to release the optimizations with GPUPI 4, but we are all smarter afterwards, right?
  7. _mat_

    Tweak book for cheats

    You misunderstood my past post. With Windows 7 GPUPI is falling back to RTC which is safe on all platforms but the EVGA SR-2 (you will be notified if it's detected). So normally no need for HPET on Windows 7 for GPUPI! On Windows 10 RS4 and before you don't need HPET if you are running Skylake or higher. All other will need HPET because RTC and RDTSC are not safe on Windows 10 Windows 10 RS5 needs HPET enabled on all platforms currently because QPC frequency was changed and GPUPI can't detect it yet!
  8. _mat_

    Tweak book for cheats

    No HPET necessary for Intel CPUs >= Skylake. Just use GPUPI 3.2 or above with Windows 10 RS4 or below. In RS5 it is currently not working because it changes the timer frequency of QPC to 10 MHz instead of the previous bootup frequency. So GPUPI 3.2 can't detect the usage of RDTSC correctly in RS5. Also no need for bcdedit, just open GPUPI, goto menu "Tools" => "Enable HPET". Reboot and you are good. You can also switch it back there. OpenCL is the main problem right now. It's a mess. The timer restrictions are necessary and good. Won't apologize for that. 😛 Btw, GPUPI 4 already uses a driver to directly access HPET and there is not need for OpenCL for CPUs anymore. Current status:
  9. _mat_

    GPUPI - SuperPI on the GPU

    To anyone wondering why GPUPI is currently crashing with the W-3175X: I am working with Intel to fix the problem. No CPU sample of course, but they get me in touch with the benchmark guys. I sincerely hope that they deliver so GPUPI can help to prove the true value of this overpriced chip. 😛
  10. _mat_

    Tweak book for cheats

    Very true, that's why I haven't disclosed anything meaningful to cheat with. I merely showed that it's possible to point out that there are problems that can not be neglected. So no tools and no instructions from my side, if that's what you meant. Ok, let's make XTU the exception. And I only disclosed after it has lost its global points and it's far more than a cheat manual (I hope). In Allen's case we are talking about fooling CB R15, right? It went exactly the way it should have been. Cheat discovered and rules changed for the better. If some results with this cheat slip through, it's because of the submission system, not because of Allen. Speaking of CB R15, this week I have successfully wrapped R11.5 and R15 with BenchMate; a project that I will make public later this year. If you are interested, follow me on FB and Twitter. Every now and then I am posting status updates about the project. That would be solution to a lot of the existing problems in the overclocking community.
  11. _mat_

    Tweak book for cheats

    Can't see anything wrong with it as well. If you find a way to cheat, publish a score and disclose the details in the thread. Mods will have to pick it up from there to make it work, no other way around it. Forcing the bot's hand, so to speak.
  12. _mat_

    Notable retired overclockers

    @tiborrr (benched like crazy back in the day! also left EKWB) @Monstru (still kicking hardware at lab501 though, but won't sub on the bot anymore)
  13. _mat_

    Intel XTU Version Library

    XTU doesn't need AVX, there are code paths inside the inner benchmark executable without AVX as well. You can also disable the AVX path via a configuration file, details in my article here.
  14. Seems like the category has to be added manually by the mods to make it work. @Leeghoofd?
  15. _mat_

    GPUPI - SuperPI on the GPU

    OpenCL support for CPUs will be deprecated in GPUPI 4 anyway, so this won't be a problem in the long run.
  16. _mat_

    GPUPI - SuperPI on the GPU

    Well, I'm not a lawyer. My personal opinion is to not give a fuck about any of that.
  17. _mat_

    GPUPI - SuperPI on the GPU

    AMD has no official links for its APP SDK, that's true. I will upload them to overclockers.at for better availability. As for Intel OpenCL runtime I have no clue what you are talking about. You are from Germany, right? Here is the latest Intel OpenCL runtime: https://registrationcenter.intel.com/en/forms/?productid=3207
  18. Looks good, I like the new responsive design. Some tables (ranking for example) still need some work by either removing unnecessary columns and/or pinning the first column and add the scrollbar to the other columns. Like this: https://zurb.com/playground/projects/responsive-tables/index.html
  19. _mat_

    HWBOT Intel XTU Benchmark Validation Regulations

    The profile should have been uploaded anyway, right? I am also getting this most of the times with the latest version of XTU and the upload works in the end.
  20. Now that XTU will have its global points removed soon, I am officially publishing my findings on this benchmark. I tried to give full insight on how to disect and uncover the security issues of XTU but also some tweaks and the possiblity to run the inner benchmark executable on its own for quick performance testing and points calculation. https://www.overclockers.at/articles/intels-xtu-analyzed This is not some kind of personal vendetta against Intel; far from it. The article's purpose is purely educational to raise awareness for benchmark security and timer reliability. This is not only about cheating, it's about the credibility of benchmarks and result databases like the bot as well. Security vulnerabilities are not taken seriously enough by benchmark developers and HWBOT in my opinion. Yes, I am going the hard way with XTU in my article of course and that's not for everyone. But there are already tools available for download that will get you ahead without any effort. So I'd like to start a discussion here on how we can improve the situation permanently. It goes without saying that any serious initiative would require a cooperation from all sides involved.
  21. I already have a Prime95 benchmark ready that scales pretty much perfectly and is very hungry for memory bandwidth. It is using the latest Prime95 version (29.4) and includes a few fixes like NUMA awareness, improved thread synronization and error checking. It's also faster than the XTU version. It seamlessly integrates into XTU so you can start XTU with an AMD CPU and just run the benchmark. It won't be far off to create a WIN32 application that runs the bench and uploads the score to HWBOT. Is this something the community would be interested in?
  22. Yes, every benchmark is breakable and nothing is 100% bulletproof. But that's not necessary, we only need to increase the cost for cheating to a level where it's easier/cheaper to just overclock and make the damn score. Overclocking has no 100,000 USD competitions, so it's not an impossible task. Regarding your benchmark and its current state of security: It took ~2 hours to break HWBOT Prime 1.0 and your bench and I've never written a single Java desktop application. it. No debugging or code changing involved, only static analysis of your executable. If you need details on how to reproduce this, just let me know and we'll chat.
  23. Congrats Allen! So much research and effort behind this awesome score!
×