Jump to content

chiliwili69

Founders [premium]
  • Content Count

    1386
  • Joined

  • Last visited

Community Reputation

415 Excellent

1 Follower

About chiliwili69

  • Rank
    Founder

Profile Information

  • Gender
    Not Telling
  • Location
    Spain

Recent Profile Visitors

1531 profile views
  1. A colleague has the chance to run the STMark on a Ryzen 7 3800X at 4.4GHz. The result was 3130!! And it can be overclocked a bit more! Remember that Intel CPUs (8700K, 9700K, 9900K) at 5.1GHz delivers around 3090 and at 5.2GHz deliver 3190. So, I think this new Ryzen line is going to be quite attractive in single-core.
  2. On the memory you have to try to go to the highest ratio in Freq/Latency. Just last week I did some research in what is available on the market (regardless of price and number of sticks 2 or 4) and make a table with the lowest latency available for each frequency, then calculated the ratio. So as said above, 3200 CL14 or 3600CL16 is the best for the money. I really don´t know is frequencies above 4000 are really worth. The problem of SiliconLottery is that they use an AVX offset of -2 for all their test. So, when they say 5.2GHz, it means 5.0 with AVX instructions (which are used by IL-2 VR). It is a kind of a trick what they play. Since IL-2 VR is very much dependent on CPU singlethread performance, the new AMD line processors (3700X, 3800X, 3900X) could be also a good choice. https://forum.il2sturmovik.com/topic/54715-is-still-intel-better-than-amd-for-single-thread/
  3. I really love the dev team for improving the clouds visuals, IMHO they are really important in a flight sim, since sky is our roads. In the last 20 years I probably have taken more than 500 flights for work, and I can still spend more than an hour in the window looking how nice are the clouds/light, etc, like the first time. Every effort to improve the IL-2 clouds and performance is really welcomed. Many thanks for this!! But, but,... I need to admit that the IL-2 clouds are not the best. Please just take a look of Battlefield1 video I made in this post (Of course, BF1 is more an arcade sim, but what the hell, both a beatiful clouds!):
  4. The problem with recorded tracks are two: 1.- Frequently they don´t work after a game update (which is understandable since new tech/features are introduced or modified for the good of IL-2) 2.- They might not fully replicate the full load of CPU/GPU since some items might been already calculated (Flying model, AI, etc). This is a minor problem anyhow. The problem with missions with Autopilot is that the AI acts differently every time. Alonzo tried that some months ago. The best solution would be that the Dev team would create a base track (always updated) to test the performance. But the problem is that they are always busy with other priorities, which is also logical since the team is always delivering what is more demanded and/or more profitable for the company (remember Companies are made to earn money) It is clear that performance is still the big issue in VR, specially with the latest devices and future to come. There are many variables out there: - Settings of individual headsets (WMR, Pitool, SteamVR, etc) - Settings of the IL-2 game - Hardware and their settings: CPUs/Overclock (Intel, AMD, frequency), GPUs, RAM freq, etc I think a performance test would be a worth item to be implemented in IL-2. I don´t think it will require a very big effort. We will be able to use it for four reasons: - To know objectively the influence every IL-2 setting. - To know the best hardware for IL-2 and IL-2 VR - To test how different VR headset behaves - To know why someone is obtaining less performance that it should be. I think this is important for the IL-2 VR users, and ultimately for the 1C/777 profits. If you agree with that you can vote here: https://forum.il2sturmovik.com/topic/39713-il-2-in-game-benchmark/
  5. I am glad you liked the Index, it is a very good option, not the best in every point (resolution, FOV, price) but good overall. Regarding performance I found the Index as the best. There is not other layers of software, just SteamVR. I still need to check the image quality improvement for SS above 100%, since 100% is already doing a big internal supersampling. So if GPU is bottleneck reduce SS.
  6. Fenris, you are absolutely right here. Sweviver didn´t say that the 8K+ has better resolution than Reverb. It was just my fault when quickly reading the below paragraph: "And well, while 8K+ looks identical in terms of SDE and already better than HP, the difference in clarity between 8KX and 8K+ is bigger than u might imagine.Simply everything in the cockpit is fully readable from a sitting position and height" He was only referring to the SDE, not to the clarity. Sorry Sweviver from here, you were not saying a false claim. I apologize. Will correct above posts.
  7. We knew from the past that there were only 4-5 threads in IL-2 VR: https://forum.il2sturmovik.com/topic/29322-measuring-rig-performance-common-baseline-for-il-2-v3010/?do=findComment&comment=499246 The red thread was the heavy one which bottleneck the whole thing. You can run now Perfmon utility to see if load is more distributed than in the past. Regarding the core clock, this is a good question. In the past we saw clearly the influence but I don´t know now. You could just create a flight record and run it in Monitor or VR at different clock frequencies and record the fps and compare. I did that in the past, look here:
  8. I was putting those pics as an example on the level of detail you will get with a 2560x1440 signal spreaded over the PimaxFOV (FOV of 8K+ and 5K+ are indentical) versus a signal of 2160x2160 spreaded over the Reverb FOV. So, the level of detail you will get with the Pimax5K+ and Pîmax8K+ will be identical since both have the same resolution. The only thing that it will be improved will be the SDE, but I was already indicating that when putting the pictures. So I think it is quite honest. It will be honest as well that Sweviver (who has Pimax8K+ and Reverb) put some facts (pictures) on the table and less words. With all of that, I love the Pimax guys. They are pulling from the VR ship and bringing leading edge tech to the domestic market, what is fabulous. But they can not make false statements. AS said before, what it matters for the detail is the signal resolution. I was putting the 360p just as an extreme example. My overall point is that "in terms of resolution accuity" the Reverb wins over the 8K+ (Sweviver made a false claim, which is not good for him and Pimax) .You can not prove the opposite with the data we have. Regardeless of the right FOV of Pimax, if you increase even more the FOV you have less and less PPD. My point was that the REverb has better PPD vertically and horizontally than 8K+. Period.
  9. I used the signal resolution (not the panel) since this is what it will be displayed after all in the panel. Bringing this to a TV/monitor example, you can watch a 360p movie in a 4K TV, so you will see the same detail than in a 360p TV. So, for details it is the signal what matters. Another thing is the SDE, you will see much less SDE in the 8K+ than in the 5K+. Regarding the vertical FOV of the Pimax I extracted those values from the rendered FOV, which is the maximum reachable: https://forum.pimaxvr.com/t/all-the-different-fovs-of-pimax-5k/16053 I was just pointing out that Sweviver was not being objective with his comment about 8K+ and Reverb. (he is a Pimax guy now, so be cautious)
  10. As it is known, IL-2 VR is very dependent on the CPU single-thread performance. We have run multiple tests and we saw that the performance (both in Monitor and VR) was correlated with a public benchmark which everybody can run in their PC (Passmark Performance Test). All those hundred of tests are collected and they produce an average value (including all sort of CPU speeds, overclocked and non-overclocked) which they public here: https://www.cpubenchmark.net/singleThread.html So they are the top performers in single-thread which is what it matters for IL-2 VR. But those values has to be seen with caution since they are just a bulk average. The interesting thing would be to see the top performers of each CPU model, not the average. To know that you can have the paid version (or use the trial in the free trial period) and you can visualize all the test (baselines) performed for every CPU model. For example the 8086K (world max is 3318) or the 9900K (world max is 3270): or Ryzen 3700X (world record 3119) or Ryzen 3900X (world record is 3197) So we can see that top performers (the ones on the right of the distribution) are very similar. Perhaps the 8086K is the one which has more in the very high range. To know the frequencies of those baselines you can go to Manage baselines section and search for specific CPU models and order them in CPU freq. In the section "Manage Baselines" you can go to advanced search an select a model (8086K for example) and request to show only baselines with more than 5.0GHz, so you can have an idea of how far each CPU can go for Overclocking:
  11. I have seen the both videos in 4K monitor and the 8KX view and detail is quite nice, much superior to the 8K+. Surprisingly Sweviver claims that the image on the 8K+ is still superior to the Reverb, which I doubt. In fact, the real resolution of the 8K+ is 2560x1440 per eye, which is the same than the Pimax5K+ (which I tested) but displayed in a 3840x2160 display (so less SDE). Just doing simplified maths (assuming same rendered FOVs than Pimax5K+, assuming full panel utilization and linear distribution): Reverb resolution per eye 2160x2160 over a smaller FOV per eye (90.2ºV, 91ºH), so 2160/90.2=23.9 PixelPerDegree Vertical and 23.7 PPD Horizontal. Pimax8K+ resolution per eye 2560x1440 over a FOV(normal) per eye(103.6ºV, 113.5ºH), so 1440/103.6=13.9 PPD Vertical and 2560/113.5=22.5 PPD Horizontal With the Pimax8KX is another story: Pimax8KX resolution per eye 3840x2160 over a FOV(normal) per eye(103.6ºV, 113.5ºH), so 2160/103.6=20.8 PPD Vertical and 3840/113.5=33.8 PPD Horizontal It would be nice if Sweviver can take some DSLR pictures of the 8KX versus Reverb for IL-2. Here I leave some pictures of Reverb vs Pimax5K+, so with the Pimax8K+ you will see same detail but with less SDE.: Reverb: Pimax5K+: Reverb(top) vs Pimax5K+(bottom):
  12. Just a note, the clarity in the Index was as before (ie good), the last update and patch didn´t affect the visual quality. It must be then something related to the new canted display thing or something specific to the Pimax software, SteamVR and IL-2. Perhaps that 4096 pixel limit of SteamVR: https://forum.pimaxvr.com/t/why-is-super-sampling-in-steam-vr-limited-to-4096x3463-always/13346
  13. With the 8086K you have 8 physical cores, 6 physical cores and if you activate Hyperthreading then every physical core creates a "virtual" core so you have 16 12 cores in total (6 physical + 6 virtual). Apparently this could create a very little bit more power , so more heat, so more temp, so less stable, so very little bit less OC. But for IL-2 VR you will not need more than 4 cores, so there is no real reason to activate Hyperthreading. Regarding AVX instructions, IL-2 make use of AVX instructions, so it will use the offset if you activate it. And those Ghz will be discounted to you clock speed. So, best is to put offset to zero.
  14. I just wanted to share one thing we tested at work with one PC, where we needed to squeeze as much as possible in single-thread performance. The PC specs are: Mobo: ATX Z390 ASUS ROG Maximus XI Apex CPU: i9 9900K CPU cooler: Noctua NH-D15S RAM: 32 GB DDR4 a 3600 MHz. Kingston HyperX Predator The PC was provided by a company who provide mounted PCs with overclock and warranty. The initial OC was 5.0 GHz for all 8 cores with 1.26V with no AVX offset and hyperthreading ON. We then explained to them that our application run just in only one core, so we might not need HT or many cores to run. Then we tested to disable HT and also disable 6 and 4 cores and increase the freq: 2 cores enabled: 5.3 GHz (both) 4 cores enabled: 5.2 GHz (all 4) All this with 1.26V and temps around 60 celsius and stress CPU app running. So, if you only use your PC for IL-2 (single-thread dependent) you can squeeze a little bit more your CPU by disabling some cores and get some more stable OC. In the case of IL-2 I think you should always maintain at least 4 cores active, since we know IL-2 VR has at least 4 threads, being one of them the heavy one. You will also save some dollars in your electricity bill. You can also use the i3-9350K (4 cores) or i5-9600K (6 cores) to do that, but those have less L3 cache. (we don´t really know the influence of L3 cache in IL-2 VR)
  15. Remember that in order to play IL-2 you have to install first an application of Steam called "WMR for SteamVR". To do that you need an account in Steam. When you want to play IL-2, you need to launch first "WMR for SteamVR" before launching IL-2 (IL-2 launch SteamVR automatically).
×
×
  • Create New...