ASRock RX 5700 XT Taichi Review

[nlyh]

What are the chances of AsRock dropping the price of this card to closer meet the competition do you reckon?

 

[Ross Thomson]

What are the chances of AsRock dropping the price of this card to closer meet the competition do you reckon?

given the looks and poor thermal/noise performance they would have to lower the price significantly below that of the red devil and nitro+

 

[EastCoastHandle]

@EastCoastHandle don't use techpowerup as reference

Everyone has recorded 34db on idle for the the 2070S and they show it at 33db on load.

And another strike with TPU.

 

[EastCoastHandle]

in what scenario? just ran mine 1146mv for the core & 950mhz on the memory on firestrike no problem at all. Memory can go higher if MPT was used.

I'm talking about the video in the OP. I'm talking about undervolting the GPU and raising the memory.

 

[Ross Thomson]

I'm talking about the video in the OP. I'm talking about undervolting the GPU and raising the memory.

I just undervolted my RD and upped the memory without any problems.

 

[EastCoastHandle]

I just undervolted my RD and upped the memory without any problems.

I see, thanks for the info.
In the video the Tiachi was UV to 1.054V or 1054mv. Which is a lot lower then what you provided at 1146mv. And, only lost 1-3 fps in games.
I'm not implying that you cannot UV a RD nor am I saying that you cannot gain a benefit from it.
What I am saying is that the Tiachi, even undervolted seems to perform exceptionally well here.

 

[Ross Thomson]

I see, thanks for the info.
In the video the Tiachi was UV to 1.054V or 1054mv. Which is a lot lower then what you provided at 1146mv. And, only lost 1-3 fps in games.
I'm not implying that you cannot UV a RD nor am I saying that you cannot gain a benefit from it.
What I am saying is that the Tiachi, even undervolted seems to perform exceptionally well here.

he dropped gpu frequency but upped memory frequency which is why there was hardly any loss in fps. My GPU was set to 2100mhz with that UV BTW.

 

[EastCoastHandle]

he dropped gpu frequency but upped memory frequency which is why there was hardly any loss in fps. My GPU was set to 2100mhz with that UV BTW.

Yeah, that's what makes it so interesting. When you UV you want to keep your OC if possible. In this case dropping OC and UV and only bumping the mem gave unnoticeable difference in performance.

 

[Ross Thomson]

memory voltage is different from core voltage.

standard set to 2100mhz GT1 71.10fps GT2 52.17 fps 13841 score 1994mhz; 56c core; 79C hotspot; 60C mem 221W
1021mv UV 2000mhz GT1 69.89 fps ; GT2 50.71 fps; 13517 score 1880 mhz; 49C core ; 65C hotspot; 58C mem; 177W
1021mv + 950 mem GT1 71.78 fps; 52.12 fps ; 13890 score; 1896 mhz; 50C core; 64C hotspot; 58C mem; 182W

Thats just a random UV and core set to 2000 mhz for the UV.



Edit I was running fire strike extreme.

 

[EastCoastHandle]

memory voltage is different from core voltage.

standard set to 2100mhz GT1 71.10fps GT2 52.17 fps 13841 score 1994mhz; 56c core; 79C hotspot; 60C mem 221W
1021mv UV 2000mhz GT1 69.89 fps ; GT2 50.71 fps; 13517 score 1880 mhz; 49C core ; 65C hotspot; 58C mem; 177W
1021mv + 950 mem GT1 71.78 fps; 52.12 fps ; 13890 score; 1896 mhz; 50C core; 64C hotspot; 58C mem; 182W

Thats just a random UV and core set to 2000 mhz for the UV.



Edit I was running fire strike extreme.

That really doesn't have anything to do with the reviewer in the OP dropping OC + UV while bumping the memory to only show 1-3 FPS difference yet bring the power and heat in line to where the Nitro is.

 

[Ross Thomson]

That really doesn't have anything to do with the reviewer in the OP dropping OC + UV while bumping the memory to only show 1-3 FPS difference yet bring the power and heat in line to where the Nitro is.

You said you hadn’t seen an RD do 930 on memory with an under volt so I showed you mine do 950 with a big undervolt. You can see the score and frame rate is virtually identical between the standard out the box settings compared to my undervolt and memory OC which is the same as what that vid shows. So it kinda has everything to do with the video.

 

[EastCoastHandle]

You said you hadn’t seen an RD do 930 on memory with an under volt so I showed you mine do 950 with a big undervolt. You can see the score and frame rate is virtually identical between the standard out the box settings compared to my undervolt and memory OC which is the same as what that vid shows. So it kinda has everything to do with the video.

To technical effect I still haven't "seen it". I may have read about similarities though .

I further elaborated to you that my post wasn't an indication that the RD wasn't "undervolt'able". Which would also cover any personal feelings you have about ram OC'ing. So to say "I haven't seen it" doesn't mean it's not possible.

In a later post I explained to you that the undervolt that I was referencing was from video in the OP. Something you advised was much higher on your RD. And, to further add, I did thank you for your post earlier.

So, to that end, to further clarify to "you" I've not seen a duplicate result of the Tiachi UV (from the OP) made on another 5700 xt. I hope that clears things up for you.