• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Discussion Nvidia Blackwell in Q1-2025

Page 160 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
reb0rn said:
limited how, is there some game which will not work?
you want let say 64GB on GPU just buy workstation card no one stopping ppl wanting to brag how their cards have lots of GPU ram
Click to expand...
I clearly said limited to 8GB in laptops, so why are you bringing workstation card to this debate?
I already showed proof above that 8GB is not enough even today.

Here is a comparison from Spiderman 2 I posted in another thread.
Models1080p RT Off
Average FPS		1440p RT Off
Average FPS		2160p RT Off
Average FPS		1080p RT Off
Minimum FPS		1440p RT Off
Minimum FPS		2160p RT Off
Minimum FPS
4060Ti 16GB vs 8GB90.8 vs 79.4 (+14.4%)63 vs 52.6 (+19.8%)33.2 vs 26.5 (+25.3%)67.4 vs 56.2 (+20%)46.9 vs 37.4 (+25.4%)20.4 vs 15.8 (+29.1%)

Models1080p RT On
Average FPS		1440p RT On
Average FPS		2160p RT On
Average FPS		1080p RT On
Minimum FPS		1440p RT On
Minimum FPS		2160p RT On
Minimum FPS
4060Ti 16GB vs 8GB49.5 vs 40.2 (+23.1%)33.1 vs 27.2 (+21.7%)17.5 vs 14.2 (+23.2%)31.7 vs 27.2 (+16.5%)21.3 vs 18.1 (+17.7%)9.9 vs 7.9 (+25.3%)
 
Last edited:
TESKATLIPOKA said:
I clearly said limited to 8GB in laptops, so why are you bringing workstation card to this debate?
I already showed proof above that 8GB is not enough even today.

Here is a comparison from Spiderman 2 I posted in another thread.
Models1080p RT Off1440p RT Off2160p RT Off1080p RT Off minimum1440p RT Off minimum2160p RT Off minimum
4060Ti 16GB vs 8GB90.8 vs 79.4 (+14.4%)63 vs 52.6 (+19.8%)33.2 vs 26.5 (+25.3%)67.4 vs 56.2 (+20%)46.9 vs 37.4 (+25.4%)20.4 vs 15.8 (+29.1%)

Models1080p RT On1440p RT On2160p RT On1080p RT On minimum1440p RT On minimum2160p RT On minimum
4060Ti 16GB vs 8GB49.5 vs 40.2 (+23.1%)33.1 vs 27.2 (+21.7%)17.5 vs 14.2 (+23.2%)31.7 vs 27.2 (+16.5%)21.3 vs 18.1 (+17.7%)9.9 vs 7.9 (+25.3%)
Click to expand...

So you think we all need 128GB cards then? What is this, 2040?!?!

😉

Personally, I like comparing the 4060Ti 16GB to the 2080Ti which is a favorable matchup. Its pretty remarkable that this much performance comes in a tidy little low power package, I don't think anyone would scoff at a 2080Ti even at this late juncture. If the 5060Ti could compare favorably to the replacement it never got, the 3070 Ti 12 or 16GB, I still think this is good and extremely useful levels of performance. Comparing it to the 3080Ti is likely going to be a bridge too far given that the Blackwell seems to be a bit of a architectural dud. Had it been on target, that might have been very possible.
 
jpiniero said:

Videocardz claims it has 3dmark 5060 Ti 16 GB benchmarks. It averages 20% faster than 4060 Ti 16 GB but still 30-40% slower than the 5070.
Click to expand...
I will add to It.
Speed Way (1440p)Steel Nomad (4K)Port Royal (1440p)Time Spy(1440p)Time Spy Extreme (2160p)Fire Strike (1080p)Fire Strike Extreme (1440p)Fire Strike Ultra (2160p)
RX 4060Ti100%100%100%100%100%100%100%100%
RX 5060Ti125.2%117.8%122.6%114.3%116%119.3%122.7%125.8%
RX 5700177.4%164.2%172.3%215.7%173.5%167.1%180%194.8%
Screenshot_7.png
Time Spy(1440p) looks wrong, RTX5070 is too fast, but others are also looking too high.
As for the 5060Ti? Hard to tell from those scores, the range is +14.3-25.8%. The not so big increase could be attributed to lower increase in TDP, supposedly only 180W, which is only 15W(9%) more, but we need gaming results.
 
MrTeal said:
Well yeah, but you had to pay $380 for your 3060 Ti. Now you'll probably be able to buy an 8GB 5060 Ti that's 14% faster in TS for the low low price of $380.
Click to expand...
14% for the same price after five years... No one is really going to even notice the difference.

It was actually 559,90 € in 24% VAT. Felt pretty bad back then but it has lasted quite long. This was one of the most over engineered 3060 Ti's out there with 250W power limit and obviously whisper quiet.
 
TESKATLIPOKA said:
Are you seriously basing the actual performance on Geekbench?
There is no reason why the gain should be any lower than what RTX 5070 has over RTX 4070( +25% at 4K TPU).
Actually, It should be even higher, because 4060Ti was BW starved.

P.S. Ok, If they don't increase TDP(TBP), then performance increase could be worse than expected
Click to expand...
I trust (older) geekbench versions for a lot of things, but measuring GPU performance isn't one of them.
TESKATLIPOKA said:
Ok, 5070 saw a 21% increase at 1440p, but don't forget 4060Ti was BW starved.
Despite having 42% more Cuda cores and much higher TDP than RTX 4060, It was only 26% higher performance at 1440p(TPU).
5060Ti has 56% higher BW thanks to GDDR7.
So It should provide higher increase than what we see with RTX 5070 unless low TDP increase.

It will be interesting to see It battle against 9060XT, although Nvidia should be faster both in Raster and RT.

Release is within a week, so we will see how It actually performs.
Click to expand...
Absolutely TERRIBLE assumptions. Even games don't scale with more cores beyond a certain point. There is quite a bit of evidence that indicates that bandwidth is NOT an issue. Lack of VRAM can be. Poorly optimized games/game engines CAN be. You'd have to show me an actual test that shows bandwidth usage for me to believe that. I've worked extensively with OpenGL and DX3-11 along with some DX12 and Vulkan stuff, and I know what I saw writing code. All of my work was done in native C or C++ btw, no engines involved.

NVIDIAs *90 series is faster, sure, but look at the power budget and increased VRAM.

I'm willing to bet (admittedly with no evidence other than profiling code) that games would see far more improvements with faster clocks over wider cores. AMD seems to be proving that so far. We'll see what happens long term.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top