Friday, January 17th 2025

NVIDIA RTX 5090 Geekbench Leak: OpenCL and Vulkan Tests Reveal True Performance Uplifts

The RTX 50-series fever continues to rage on, with independent reviews for the RTX 5080 and RTX 5090 dropping towards the end of this month. That does not stop benchmarks from leaking out, unsurprisingly, and a recent lineup of Geekbench listings have revealed the raw performance uplifts that can be expected from NVIDIA's next generation GeForce flagship. A sizeable chunk of the tech community was certainly rather disappointed with NVIDIA's reliance on AI-powered frame generation for much of the claimed improvements in gaming. Now, it appears we can finally figure out how much raw improvement NVIDIA was able to squeeze out with consumer Blackwell, and the numbers, for the most part, appear decent enough.

Starting off with the OpenCL tests, the highest score that we have seen so far from the RTX 5090 puts it around 367,000 points, which marks an acceptable jump from the RTX 4090, which manages around 317,000 points according to Geekbench's official average data. Of course, there are a plethora of cards that may easily exceed the average scores, which must be kept in mind. That said, we are not aware of the details of the RTX 5090 that was tested, so pitting it against average scores does seem fair. Moving to Vulkan, the performance uplift is much more satisfying, with the RTX 5090 managing a minimum of 331,000 points and a maximum of around 360,000 points, compared to the RTX 4090's 262,000 - a sizeable 37% improvement at the highest end. Once again, we are comparing the best results posted so far against last year's averages, so expect slightly more modest gains in the real world. Once more reviews start appearing after the embargo lifts, the improvement figures should become much more reliable.
Sources: BenchLeaks, VideoCardz
Add your own comment

33 Comments on NVIDIA RTX 5090 Geekbench Leak: OpenCL and Vulkan Tests Reveal True Performance Uplifts

#26
razaron
Hopefully that's unoptimised drivers...

Factoring in core count and clock increases, that slightly less to equal IPC. You'd expect at least 5% ipc improvements even with bad gen uplifts.

Nvidia getting rid of 40xx stock early makes more sense now. Due to that I'll probably still be interested in a 5080/90 if I find them at MSRP, since I'm after AV1 and the alternatives aren't worth the upgrade just for a codec.

Also, fun thought excercise. 8800 GT was $350 and RTX 4090 is apparently >60x faster. So RTX 5090 should cost 28k (42k factoring inflation). Thank god bullshit price performance increases are only a past few gen phenomena.
Posted on Reply
#27
werda
Recent GPU launches seems to higlight the importance of semiconductor manufacturers on IPC gains. If TSMC nodes are not improving fast and cheap designers are simply unable to design their way into big IPC gains. Latest launches made me realize TSMC is even charging less than it could have given how vital they are for meaningful efficiency gains.
Posted on Reply
#28
Vya Domus
AnotherReaderNvidia has invested in OpenCL support as well.
No they haven't. They kept their GPUs on OpenCL 1.2 for years and years and that effectively killed it, they were the dominant GPU manufacturer but they only offered basic API support, they made sure OpenCL had no chance to gain traction. It was a very smart move for their interests but it was clearly anti competitive and anti consumer, no doubt about it.

If Nvidia invested in OpenCL support CUDA would no longer exist today lol.
Posted on Reply
#29
Zach_01
b1k3rdudeNO, it is Not!, its not how economics works, or has ever worked. By that bastardised logic, GPUs will just keep increasing in price to where no one will be able to buy them and the market segment will collapse. You like so many others has let themselves be fooled by the gaslighting enshitification that nvidia has been pushing. Would you consider 3000 or 4000 (the price of a used car/van) for a consumer GPU is justifiable let alone sustainable..? If jacket-boy and his cohorts keep this up, its going to be the beginning of then end. Look at Intel, they fcuked the pooch over a prolonged period badly enough that an external investment corp are looking at buying out the whole co.
I agree that nVidia market practices towards gamers and even towards prosumers on some level are at least questionable (a lot worse really) we cannot directly compare the company to Intel.
The danger of nVidia to be in the place of where Intel is now is slimmer than a laser cut.
Intel started to lack innovation and eventually competitiveness on all levels of CPU products while charging absurd prices.

The main difference with nVidia is that their products and their architecture are selling like the sweetest hot cakes that cure cancer on the professional level (AI/ML). They are not getting out of business anytime soon. Quite the opposite.
Even if the majority of gamer customers will never buy nVidia again.

So it’s not apples to apples comparison. Far from it
Posted on Reply
#30
Rightness_1
This is going to be the worse gen-on-gen update ever. +-15% for most "consumer" SKU's.

I was so excited about upgrading my ancient 2070, but this 50x0 series is not looking good at all. The perf uptick from a 4070Ti Super to the "new" 5070Ti is about 5-10% at best.

I also think 16GB of VRAM is not enough, which rules the 5080 for me.

I don't want to spend $1000 a year for drip drip drip cards.
Posted on Reply
#31
windwhirl
Rightness_1I was so excited about upgrading my ancient 2070
Eh, I'd say a 5000 series would be a worthwhile upgrade if you're still rocking a 2070, but
Rightness_1I also think 16GB of VRAM is not enough, which rules the 5080 for me.
If 16 GB of VRAM isn't enough for you, then you're dead in the water because more VRAM than that is either the 3080 Ti 20 GB variant (I didn't even know/remember that existed) or x090 cards.
Posted on Reply
#32
Godrilla
What if Nvidia's goal is to make fe edition slightly better than last gen only to make the aib cards that will probably be more expensive more attractive and performant in comparison ?

Paper launch fe edition and flood the market with aib cards that are more expensive but potentially clock higher.
Bait and switch. lol.
Posted on Reply
#33
Rightness_1
windwhirlEh, I'd say a 5000 series would be a worthwhile upgrade if you're still rocking a 2070, but


If 16 GB of VRAM isn't enough for you, then you're dead in the water because more VRAM than that is either the 3080 Ti 20 GB variant (I didn't even know/remember that existed) or x090 cards.
In one way yes, you're right, but I do not have the money for a $1000 card every year. I need 3 years out of this card, and I feel that when it is only 15% faster than a 4070Ti Super, then where is the value?

There are games right now that can't run at 60 fps at 4K. I know there are games coming this year that will struggle to run fast on what the 5070ti is going to be, so it feels like a bad investment right now.

I have no issue with a xx7x card only having 16GB of VRAM this year, but the xx8x only having 16 is a problem to me, as it it offered 24GB, it would have been a day 1 purchase, but as it stands, 16GB offers no longevity whatsoever, a year at best.
Posted on Reply
Add your own comment
Nov 20th, 2026 19:28 PST change timezone

New Forum Posts

Popular Reviews

TPU on YouTube

Controversial News Posts