• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

The Radeon RX9070XT / RX9070 Owners Thread

Kirkatron said:
They are an outlier. 2nd place is 8030 or something. They must have used cool air or are some kind of OC genius.

I need 8048 to top my table. I don't stand a chance. I'm nearly 200 points off!
Click to expand...
No OC genius, these cards are super sensitive to temperature so I just put my pc with the case open next to an open window on a cold morning.

I also highly recommend trying to do negative offset. My highest score which is 8227 was done with a -140 core offset.
 
Last edited:
WARburton said:
hahaha god no, but are the bad 9070xt's when benching....

show me a gaming related test and i will play for an hour to see what i can get
Click to expand...
I'd hope an XT at 70fps was game stable. They are probably just stock XT scores around that region. My game stable XT bench in Steel Nomad is 78.5fps. Going to try stupid settings at some point to push the score.
 
Kirkatron said:
I'd hope an XT at 70fps was game stable. They are probably just stock XT scores around that region. My game stable XT bench in Steel Nomad is 78.5fps. Going to try stupid settings at some point to push the score.
Click to expand...

yes the good ones would be, but there is a few in this tread that are struggling to got 7400 points when maxed out, for a none xt to do 7000 when maxed out its good buddy.
and that would all depend on the game, as they all put different loads on the gpu. there is no one box fits all when overclocking

you have a notro+ how dose your bench? at game settings
 
Last edited:
Kirkatron said:
I'd hope an XT at 70fps was game stable. They are probably just stock XT scores around that region. My game stable XT bench in Steel Nomad is 78.5fps. Going to try stupid settings at some point to push the score.
Click to expand...

My game stable bench (playing a game for an hour without a crash) is 74fps (7445) at Steel Nomad but if I push mine I get 77fps but some games don’t even start at that level of UV.
 
Last edited:
WARburton said:
yes the good ones would be, but there is a few in this tread that are struggling to got 7400 points when maxed out, for a none xt to do 7000 when maxed out its good buddy.
and that would all depend on the game, as they all put different loads on the gpu. there is no one box fits all when overclocking

you have a notro+ how dose your bench? at game settings
Click to expand...
I put the rough game bench in that message, but it is 7866 78.7fps to be exact.

I'm just shy of the 8000 score when maxed out. 7950, 79.5fps. So I don't really gain much by making it unstable. Done a few runs in the last few mins and got to 7950. Can't quite crack 8000.
 
Kirkatron said:
I put the rough game bench in that message, but it is 7866 78.7fps to be exact.

I'm just shy of the 8000 score when maxed out. 7950, 79.5fps. So I don't really gain much by making it unstable. Done a few runs in the last few mins and got to 7950. Can't quite crack 8000.
Click to expand...
When im game stable its about 6800 66fps, as to be expected.

pushing 8000 is a good score
 
Fdk5elt.png


Made it into the top 10 for 13700K with 9070 XT. Although technically 2nd as the 8 places above me are the same person. I don't think i can get any more out of this GPU.
 
Last edited:
Kirkatron said:
Fdk5elt.png


Made it into the top 10 for 13700K with 9070 XT. Although technically 2nd as the 8 places above me are the same person. I don't think i can get any more out of this GPU.
Click to expand...


Assuming you're using the higher scoring Vulkan API that score is 45% higher than an RTX 5070, 11% higher than an RTX 5070 Ti, and an RTX 5080 is 10% faster.

Its a bit nuts to think AMD have gone from what they were in RT to this in one generation, I liken this to a replacement for the RX 7800 XT, this is 75% faster in DXR.

www.guru3d.com

GeForce RTX 5070 Founders Edition review (Page 24)

We review and benchmark the GeForce RTX 5070 Founders Edition. The fourth product in the GeForce RTX 5000 range is built using a 5 nm manufacturing process and runs at 2165 MHz but can boost to around 2510 MHz under load conditions. Priced at $549 and 649 EURO will NVIDIA offer you enough value...
www.guru3d.com www.guru3d.com
 
Last edited:
humbug said:
Assuming you're using the higher scoring Vulkan API that score is 45% higher than an RTX 5070, 11% higher than an RTX 5070 Ti, and an RTX 5080 is 10% faster.

Its a bit nuts to think AMD have gone from what they were in RT to this in one generation, I liken this to a replacement for the RX 7800 XT, this is 75% faster in DXR.

www.guru3d.com

GeForce RTX 5070 Founders Edition review (Page 24)

We review and benchmark the GeForce RTX 5070 Founders Edition. The fourth product in the GeForce RTX 5000 range is built using a 5 nm manufacturing process and runs at 2165 MHz but can boost to around 2510 MHz under load conditions. Priced at $549 and 649 EURO will NVIDIA offer you enough value...
www.guru3d.com www.guru3d.com
Click to expand...
That was DX12 version. I've not tried the Vulkan one.

Here is the score and details:
https://www.3dmark.com/sn/5286621
 
Last edited:
humbug said:
Assuming you're using the higher scoring Vulkan API that score is 45% higher than an RTX 5070, 11% higher than an RTX 5070 Ti, and an RTX 5080 is 10% faster.

Its a bit nuts to think AMD have gone from what they were in RT to this in one generation, I liken this to a replacement for the RX 7800 XT, this is 75% faster in DXR.

www.guru3d.com

GeForce RTX 5070 Founders Edition review (Page 24)

We review and benchmark the GeForce RTX 5070 Founders Edition. The fourth product in the GeForce RTX 5000 range is built using a 5 nm manufacturing process and runs at 2165 MHz but can boost to around 2510 MHz under load conditions. Priced at $549 and 649 EURO will NVIDIA offer you enough value...
www.guru3d.com www.guru3d.com
Click to expand...

Steel Nomad is pure raster though isn't it? There is no RT in that benchmark
 
Apocryphal said:
No OC genius, these cards are super sensitive to temperature so I just put my pc with the case open next to an open window on a cold morning.

I also highly recommend trying to do negative offset. My highest score which is 8227 was done with a -140 core offset.
Click to expand...
Your score is ridiculous and with a 7700 too!

I have done negative offset to gain stability.
What settings did you use to get 8227, and for your 5700x3d score, as your average GPU clock was quite a lot higher than mine and we were only 6 degrees Celsius difference on GPU temp.
 
Last edited:
Kirkatron said:
Your score is ridiculous and with a 7700 too!

I have done negative offset to gain stability.
What settings did you use to get 8227, and for your 5700x3d score, as your average GPU clock was quite a lot higher than mine and we were only 6 degrees Celsius difference on GPU temp.
Click to expand...

For the R7 7700 score I had -125mV and -140 core clock offset and for the 5700X3D I am not entirely sure, but I would guess around -110-115mV and -150-200 on the core clock.

And 6C might not seem a lot, but I would say that would make a difference of atleast 50points in Steel Nomad which is quite a lot.

And since I saw someone mentioning increasing the base clock for the 5700X3D I can mention that I was running with a base clock for 101 for it. I dare not raise it higher since it is my wifes PC and I just wanted to test my GPU on her system to see how it was running.
 
You must log in or register to reply here.
Back
Top Bottom