wanna to hear something āfunnyā?
I did the bench on my i9-285k / RTX3090 combo with TVAI 7.0.
and my results are lower than yours.
I only get 30% of the results i got in the TVAI V6.0 benchmarks before.
Especially with gaia 2x i used to get 6.22 fps. But with 7.0 its just 2.41 fps.
Yes, i still use gaia regulary because for me it gives best results on poor sd footage.
hmm, strange, after a system restart, the V7 results are pretty close to the V6 results again.
do i need to do a benchmark each time i want to render something,
so i know TVAI=V7 is in full power mode?
however i keep testing the v7 and already know that this unbelieveable long starlight rendering times, in my case 18h9min for a 2min footage, will not make starlight an option to me.
My benchmarks have not been significantly different for most release versions from v 3.x on (there have been a few problematic builds along the way). This consistency is probably the flip side of benchmarks being slower for AMD GPUs than for Nvidia.
have you ever looked at what your gpu is doing when you get these terrible results? I ask because Iāve gotten terrible results before and noticed what looked like memory leak where it fills up the entire GPU memory as it is loading the model and then starts using 4-8gm of the system ram reducing efficiency by 200-400%. do you have it set to use 100% memory might try reducing that just ran into this today so made me think of it.
yes, of course. before i do benchmarks i always make sure that cpu and gpu are in idle mode.
therefore i was so surprised and restarted the system. since then tvai 7 performs normal.
What are you trying to say? Last year it was the Intel CPUs that were burning up. If they both have the same potential to burn up, then AMD is still better because it preforms better.
You knowā¦ I really donāt like putting you on blast, but you just makes too easy for me. Seriously!
This AMDumbā¦ againā¦ CANNOT be corrected with a BIOS revision unlike the 13- 14-gen series owners that had rapidly applied ALL updates BEFORE any oxidations seared in.
But in the video says it should be fixed with the BIOS updateājust like what Intel didā¦ Itās a very similar. Only you have said it cannot be fixed with a BIOS update. Where do you get the critical information to make such an important claim? Please share.
No, I did not state thatā¦ SMH! I saidā¦ it CANNOT BE CORRECTED (and that OPINION is not confirmed in the video)! The two words are not the same.
Resolved/Fixed = dormant/masking/circumventing, ect (that will RETURN again but it could be even worse the next timeā¦ by breaking other stuff later) the issue!
Addressed = Alpha/Beta/ect. releases that mostly states; "we are vetting the issue(s), āwe might have a stack for trialā, ect.
Corrected = 98.9898% of time means that issue will never rise-up again!
Intel never removed any negative posts in their forums about the 13- & 14-gen (and/or threatened shilled social handlers outlets) like AMD is doing at this second, ect. AMD still refuses to comment to its 9000 series diaster since reports in March (technically, since ~Dec/2024 for its engineering samples silicon before its release in 3/25).
Suā¦ againā¦ is just shamedā¦ and should be. Itās Suās diaster Cell processor (PS3), ect. all over again.
Thanks for that update. I have been interested in the Asrock/AMD issues because quite a few of the motherboards I have shortlisted are Asrock ones and this is now the most recent update Iāve seen.
Nevertheless, based on all the benchmarks I have seen, the 9950X3D is better than Intelās current flagship 285K CPU in gaming, productivity and power efficiency. Unfortunately for Intel at this time their latest 285K CPU even falls behind their previous generation 14900K in many benchmarks, which also trails the 9950X3D.
While the 14900K and 285K are cheaper than the 9950X3D, they donāt fare much better against the cheaper 9800X3D either. The only reason for me to go with intel at this point would be if I opted for the 14900K as it would be compatible with my current LGA1700 motherboard.
But like I said, I have no particular preference towards the AMD or Intel brands themselves. I will go with whatever product best suits my requirements.
Remember, synthetic benchmarks, especially from shilled [or just haters/bashers] outlets, DOES NOT reveal the full-story (not even close) and these metrics should ONLY be considered as a quick first-look, not as an overall deciding measure for systems stabilityā¦ a peace of mind.
I donāt think the 3D variant will benefit you much for TVAI, but if you want it for other applications, that makes sense. Either way, it should be a good option for you. Gamers Nexus sets a pretty high bar for review testing and quality.
