• Hello Guest!

    We have recently updated our Site Policies regarding the use of Non Commercial content within Paid Content posts. Please read the new policy here.

    An offical announcement about this new policy can be read on our Discord.

    ~The VaMHub Moderation Team
  • Hello Guest!

    We posted an announcment regarding upcoming changes to Paid Content submissions.

    Please see this thread for more information.

Benchmark Result Discussion

4080 is faster by about 15% than 3090Ti. People hated it cuz it was priced pretty high, not because it's perfomance. I mean, non Ti-version of high-end card beats down the Ti version of previous gen 'top tier'.
I'm pretty sure even 4070Ti is faster than 3090Ti - no much, probably within a margin about 5-10%.
And it's just considering a 'raw power', no rtx\dlss technlogy\hardware updates.

7900XTX is almost equal to 4080. But no fancy Nvidia stuff, rtx performance is closer to 30xx cards than current Nvidia gen.
As for VaM, there was someone who tested it vs 4080 [with 13900k cpu if i recall] and 4080 was performing better.
 
4080 is faster by about 15% than 3090Ti. People hated it cuz it was priced pretty high, not because it's perfomance. I mean, non Ti-version of high-end card beats down the Ti version of previous gen 'top tier'.
I'm pretty sure even 4070Ti is faster than 3090Ti - no much, probably within a margin about 5-10%.
And it's just considering a 'raw power', no rtx\dlss technlogy\hardware updates.

7900XTX is almost equal to 4080. But no fancy Nvidia stuff, rtx performance is closer to 30xx cards than current Nvidia gen.
As for VaM, there was someone who tested it vs 4080 [with 13900k cpu if i recall] and 4080 was performing better.

I also wait for an 7900 xtx bench like hell, but except this anecdotical evidence - that “sumbody tried and aint worth it..” isnt really anything. Yes in VR 7xxx suxx atm, but I use it in pancake mode - aka desktop - and the 6xxx put up a pretty good fight in prev gen. So chances are that it may do it again. Sadly I cant try cuz I have a custom loop sff desktop, and exchanging hw is as painful as possible (did it on purpose :D). But if there are anyone out there please hear us here and throw some results!
peace
 
I am planning to get a 7900XTX myself but probably not until around June.

The 'problem' we face here trying to get an idea of how the 7900XTX might do in VAM is having no similar references from most reviews of the card out there. Most games being tested in reviews are modern (or not that old, generally-speaking maybe up to Witcher 3 is one of the oldest commonly recurring game being tested to this day but that's like 2015'ish, not that bad). But more importantly, most of them are not Unity games. It's tough to gauge then how the 7900XT, or XTX would do in VAM. Then again VAM is using a pretty old Unity engine version, too. I'm not even sure if there's any known games that happened to be using the same version of Unity as VAM does right now.

At the moment finding benchmarks of the 7900XTX where games like...

- Subnautica
- Rust
- Cities: Skylines
- Fe

... are being tested, would be at least some indication of the XTX's performance (next to other cards, for those games in particular).

I can't wait to get mine but the price makes me have to save up some extra pennies, it's a pretty hefty purchase for a single hardware component.
 
So right now my current system is
AMD R5-3600 @ 4.0ghz
32GB DDR4
AMD Radeon 5700 XT (Sapphire Nitro +)

I'm wanting to upgrade either to the R7-5800X or R7-5800X3D

I've looked over the benchmark results of the X3D and it seems to be one hell of a processor for VAM, I'm just curious though if it's really going to be worth paying the extra $100 to go with the X3D over the standard 5800. Especially if I decide to leave the same video card I currently have?


And lastly, I'm aware this is probably a very stupid question, but is the jump from the standard R5-3600 to either of those processors really going to be that noticeable in VR for VAM in its current state?

I'm just worried I'll end up dropping this money and VAM will still be...well.. running like VAM lol
 
So right now my current system is
AMD R5-3600 @ 4.0ghz
32GB DDR4
AMD Radeon 5700 XT (Sapphire Nitro +)

I'm wanting to upgrade either to the R7-5800X or R7-5800X3D

I've looked over the benchmark results of the X3D and it seems to be one hell of a processor for VAM, I'm just curious though if it's really going to be worth paying the extra $100 to go with the X3D over the standard 5800. Especially if I decide to leave the same video card I currently have?


And lastly, I'm aware this is probably a very stupid question, but is the jump from the standard R5-3600 to either of those processors really going to be that noticeable in VR for VAM in its current state?

I'm just worried I'll end up dropping this money and VAM will still be...well.. running like VAM lol
I would recommend the x3d variant definitely. Average fps might not jump much, but lows going to be much better, hence overall smoother. I have a 5600x and saw very similar 8 core results. Only the x3d was in the same ballpark as the intel alder lake.
GPU wise you can try amd builtin RSR (upscale in adrenalin) and lowering tesselation (in adrenalin aswell).
 
I would recommend the x3d variant definitely. Average fps might not jump much, but lows going to be much better, hence overall smoother. I have a 5600x and saw very similar 8 core results. Only the x3d was in the same ballpark as the intel alder lake.
GPU wise you can try amd builtin RSR (upscale in adrenalin) and lowering tesselation (in adrenalin aswell).

Thanks for the suggestion, I honestly didn't even think about how much the 1% lows would improve which is arguably the very thing you wanna improve upon in a game like VAM, I'm definitely leaning towards just getting the X3D from what I've seen/read so far, and plus it will make for a nice final upgrade for this system until I finally decide to build an entirely new rig years down the line.

Also your suggestions with AMD's RSR/Upscaler settings, Do those settings actually apply in VR mode? I was under the impression that things like AMD Boost and most graphics driver settings don't apply inside VR.
 
Thanks for the suggestion, I honestly didn't even think about how much the 1% lows would improve which is arguably the very thing you wanna improve upon in a game like VAM, I'm definitely leaning towards just getting the X3D from what I've seen/read so far, and plus it will make for a nice final upgrade for this system until I finally decide to build an entirely new rig years down the line.

Also your suggestions with AMD's RSR/Upscaler settings, Do those settings actually apply in VR mode? I was under the impression that things like AMD Boost and most graphics driver settings don't apply inside VR.
Good question and I cant say for sure - as said I use desktop mode - but tessellation settings I think are applied and fsr is open source and there were a useable mod for vr way before rsr appeared in adrenalin. In your case I would give a try for rsr since its quite easy, then go look for fsr vr mode if that didnt work out.
 
I haven't done much tweaking except for ram timings, but using Process Lasso to run VAM on the 3d cache ccd. I went with a 7950x3d because I need the extra cores for multi-tasking.

I kind of feel like I should be getting more performance by looking at some of the other results but maybe I'm not reading things right. Trying to decide if I should dive into the bios more.

Benchmark-20230424-090704.png
 
I haven't done much tweaking except for ram timings, but using Process Lasso to run VAM on the 3d cache ccd. I went with a 7950x3d because I need the extra cores for multi-tasking.

I kind of feel like I should be getting more performance by looking at some of the other results but maybe I'm not reading things right. Trying to decide if I should dive into the bios more.

View attachment 237802
Is that on you main game folder or a fresh install folder? If its on you main folder with lots of vars installed, its about right. Ofcourse with PBO and Curve optimiser enabled in bios, makes a difference too.
 
Are you using a fresh install file of VAM? And Expo on? My Ram is 6000mhz CL30 Gskill, Resize bar on, PBO enabled, Curve Optomizer on and set to -30.

Hello. What is Expo and where to set it? Where to set Resize Bar and what is PBO and Curve optimizer (and where to set it).
What ram do you have ?

Im on way to choose between 7950x3d vs 13900KS. Meshed suggest that in vam 1.x 3d cache would be better than 6GH in 13900KS but VAM 2.x has diffrent job system and raw power may be better. Thouth i would prefere AMD couse of its low TDP and thus quiter cooling system :)
 
Last edited:
I think your addonPackages folder is too big, you can move files to another folder, and get huge improvement.
The size of addonPackages does not matter for fps ( when all is loaded) what matter every time is quantity of all morphs (so the more vars you have the more morphs couse of doubling and combined morpsh) so with large number (like my 2 year 800GB VAM ) it cut almost 100FPS just couse of that :)

So to get proper to compare result benchmark should be made on clean install just with benchmark loaded.

Heres detailed thread with official Meshed statement: https://hub.virtamate.com/threads/benchmark-clear-install-vs-many-vars.37256/#post-106140
 
Just re-run benchmark, for the first time at version 4, with updated VaM and...
It went slightly better than V3 with 1.20.77. My physics time at Baseline 3 got reduced by 0.05ms oO
So, even if current benchmark 'should be comparable' with older versions, VaM updates gave us some boost.


Seems like yours GPU is either power limited, or not performing in full potential.

Also seems like Intel finally got some competition, 7950X3D seems to have better low1% in every single test.
But it also so unstable. It loosing in averages, in very single test again, by 10-20%

View attachment 223810
A note: I'm using now 4 sticks [yeah i'm dumb and cheap] of DDR5 5600.
I actually bought 2 sticks of 6000 a few months back, and it gave me 1-3% boost, but i returned it cuz was a bit short of cash. It was totally a good decision tho. At the time i returned them, they were cheaper by 100$ to buy already, 2 weeks after i bought them lol

Is it 13900K or KS ? (does vam know if its k or ks ?)) ?
What GPU do you have ? (how to check if its power limited) ?
Which cooler do you use for CPU?
 
Hello. What is Expo and where to set it? Where to set Resize Bar and what is PBO and Curve optimizer (and where to set it).
What ram do you have ?

Im on way to choose between 7950x3d vs 13900KS. Meshed suggest that in vam 1.x 3d cache would be better than 6GH in 13900KS but VAM 2.x has diffrent job system and raw power may be better. Thouth i would prefere AMD couse of its low TDP and thus quiter cooling system :)
It's all in you motherboard bios in different places. 2x8gb gskill trident-z neo 6000mhz cl30
 
Is it 13900K or KS ? (does vam know if its k or ks ?)) ?
What GPU do you have ? (how to check if its power limited) ?
Which cooler do you use for CPU?
Yeah VaM should be able to recognize KS. I'm just on K, cooled with NZXT Kraken Z73.
Actually i never saw anyone with KS in this thread, but since the benchmark can tell if it's K, KF, or mobile, why it shouldn't show KS too.

As for GPU, I'm at watercooled 4090 [Aourus], with it's stock settings for benchmarks.
Basically if you won't power limit it yourself, and yours PSU can handle it, it won't be power limited.
I could tell\guess from some other people benchmarks that they were power limiting theirs 4090 - cuz of lower GPU heavy tests results than mine.

I'm all over Intel guy.
But if you going to make a brand new build right now, 7000-3d cache is faster than 13900k even in VaM 1.x right now.
If u know how to configure it.
And if you don't mind dealing with AMD. Just saw some news a few days ago that 7900-3d melted some motherboards latelly:
 
Is that on you main game folder or a fresh install folder? If its on you main folder with lots of vars installed, its about right. Ofcourse with PBO and Curve optimiser enabled in bios, makes a difference too.
It's a clean install... although I had just updated the bios so didn't realise my ram setting had been set back to stock. Think I'll wait till gamers nexus report before playing with the ram timing and tweaks.
 
Hey, are you still liking the 13900K + 4070Ti combo? I'm thinking about upgrading to something similar, although I'll probably stick to the 13700K. From your benchmarks it looks like a good setup without being insanely expensive. Thanks!
Yeah, it's running pretty smooth. I did do a little light overclocking, but nothing major - generally very happy with the build.
 
Yeah VaM should be able to recognize KS. I'm just on K, cooled with NZXT Kraken Z73.
Actually i never saw anyone with KS in this thread, but since the benchmark can tell if it's K, KF, or mobile, why it shouldn't show KS too.

As for GPU, I'm at watercooled 4090 [Aourus], with it's stock settings for benchmarks.
Basically if you won't power limit it yourself, and yours PSU can handle it, it won't be power limited.
I could tell\guess from some other people benchmarks that they were power limiting theirs 4090 - cuz of lower GPU heavy tests results than mine.

I'm all over Intel guy.
But if you going to make a brand new build right now, 7000-3d cache is faster than 13900k even in VaM 1.x right now.
If u know how to configure it.
And if you don't mind dealing with AMD. Just saw some news a few days ago that 7900-3d melted some motherboards latelly:
I was going to buy 7950x3d (need cores for work) i play very seldom, mainly play with vam :) so both gpu and cpu was vam driven and see you have about 15% better bench result then for 7950 so i get little down , but as it Amd draw 30% less power and in other games is better it seem that will be my next cpu (after 15year of intel domination ) :D
 
I just enable XMP, it run at 6400MHz CL32. I tried disabling SMT and get higher CPU utilization but it produce worse result as follow
Benchmark-20230507-093852.png
 
I just enable XMP, it run at 6400MHz CL32. I tried disabling SMT and get higher CPU utilization but it produce worse result as followView attachment 242502
Ok. Interested to see how ram scales with this. I just have the 6000mhz cl30, but also enabled pbo and curve optimizer -30 in bios. Also check you soc voltage. Asus and other motherboards likes to push high voltage and burn out these processors.

Benchmark-20230411-233848.png
 
Last edited:
Back
Top Bottom