Can someone post a benchmark of a 5070, 5080, or 5090 Mobile GPU.
Im looking for a portable laptop, with the occasional topaz use, but want to know what the results are. So i can scope how much i wanna pay, and if if its worth paying the extra or not.
I know the tensor cores vs desktop will be less. But some of us dont want a desktop and want to use it for occasional upscales (old cartoons, a movie once in a while ) etc… i currently use topaz on my laptop with 3050ti and im content waiting if i need to.
In the 5 months you have been posting on these forums your worthwhile contributions have been Zero.
Snarky comments and a false belief that you are far superior to anyone else seems to the common trait with all your posts.
If you have nothing worthwhile to contribute lay off the keyboard.
i actually use it on a 3050ti Intel i9 z13 notebook and its pretty good… takes about 1-2 days to upscale 2 hour 1080 movie to 4k HDR … im happy with that. Just time for a new notebook, so i figure maybe it could get done in 12 hours… on a desktop what.. maybe 6-8 hours? not much of a savings in my opinion. Its not like i do video editing full time
Even if it were Borderline, it is not his intention.
Borderline is a behavioral disorder and not willfulness.
Either don’t react or you justify yourself but without attacking.
I noticed that the Hyperion benchmark for 1920x1080 is about half the speed I would expect. That’s based on two comparisons:
Apple M3 Ultra and M4 Max are both much faster for Hyperion at 1920x1080 (at least by 1.5 x). Surely that shouldn’t be the case?
and
Your Hyperion benchmark for 3840x2160 at just over 20 fps should extrapolate to around 80 fps at 1920x1080, given 4 x the number of pixels. Even after allowances for processing inefficiencies, I’d expect that you and others with similar hardware should be getting much more than about 39 fps at 1080.
If this is just a benchmark issue and in reality you’re getting much higher fps for Hyperion at 1080, then sorry to bother you. Otherwise, perhaps Topaz need to be made aware?
I don’t have an answer for you as I don’t use Hyperion, but happy to run a test this week and will let you know if “real world” performance deviates from the benchmark.
GPU is SUPRIM LIQUID SOC model left in “silent BIOS” for safety reasons (TDP limit at 575W)
RAMs are 2x32GB, DDR5 6400, CL32-39-39-102 (XMP profile)
(no OC on anything, default BIOS)
TOPAZ installed on “Kingston FURY Renegade NVMe 4TB” (think its 4gen, PCIE lanes are not shared, runs on chipset ones)
BEFORE updating to 5000 optimized build (Topaz6.1.3)
I agree. With my old PC of i7-10700 and RTX 4060 Ti (16GB) for hyperion i was at benchmark 12.58 fps and 11fps real-life performance.
With my new 285K and 5090RTX i am at benchmark 22.99 fps and real-life 17fps.
I have no idea why and was definitely expecting way bigger jump or improvement.
(to add an insult to injury, i am even slower than most people here with similar setup and i doubt its just my choice of GPU silent BIOS, shoudnt make such a huge difference anyway)
It seems that TVAI does not perform very well with 285K for some models from what I have seen. One of them being Hyperion.
I also wonder about RAM speeds because in theory 6400CL32 doesn’t sound too slow but all your results which involve 4X scaling are pretty low which indicates a RAM speed bottleneck. Maybe you can run AIDA64 RAM speed test to see how that performs (free version will obfuscate some test results in RAM speed test dialog but u can run each test separate from menu to get those results too).
Sometimes default RAM sub-timings are really awful and bring down performance a lot.
Here are my results with a Ryzen 7950 + 5090 and optimized 6200CL30 RAM:
Probably needs more optimization from Topaz.