AMD: Speculation, Rumors, and Discussion (Archive)

Status
Not open for further replies.
Razor1 said:
http://videocardz.com/61225/amd-radeon-rx-480-rumors-part-2
3dmark scores with pictures
Click to expand...

Here's FS extreme on my 24/7 OC 970 for comparison

1fpj5t.png


If that's at stock clocks, that's very impressive.

Edit: FS Ultra result: https://abload.de/img/2erjx1.png
 
Last edited:
Gipsel said:
If the number of memory chips as well as the bus width stay constant, not much changes irrespective of the capacity (the changes are tiny). If you add more chips to increase capacity, yes in that case the power consumption rise. But even then, if you keep the bus width constant, it is actually much less then linear. If you put two GDDR5 chips on a single 32bit Controller port (i.e. run them in clamshell mode where each chip provides 16 bits for each transfer), they won't consume twice as much as a single chip operating in 32bit mode. Last time I checked the power consumption in clamshell mode was something like 60-70% of the maximum when operating in 32bit mode.
Click to expand...
Wait, clamshell mode generally consumes less power? Or are you saying it's less than twice the power from the chips? Very neat detail!
 
Some people are saying that the 54xx result was under older driver while the 58xx result is from a newer one which still doesn't detect the card properly. So maybe there's some gas still left in the tank and it can match 390X.

Looks a decent card especially if AMD have sorted out their lower resolution troubles and clockspeed headroom.
 
CSI PC said:
Sorry if this seems pedantic. but by that logic you might as well say there is 0W consumption for the GDDR5 VRAM modules.
If there is zero difference between 4GB and 8GB, there is also negligible difference between none and 4GB (ok some but not much - yeah being too pedantic I agree as deliberately ignoring the interface :) ).
Click to expand...
If you ignore the interface then you arrive at some funky results for a zero gigabytes RAM capacity; you're not going to have any data traffic across a host interface with no DRAM attached to it, so host interface would not draw much power. So 0 GB RAM would lead to much less total power draw versus 4 or 8 GB capacity. :p 4GB vs. 8GB would not be that big a difference on the other hand.

As DRAM is basically a grid of extremely tiny capacitors (with some support logic attached, obvs), you're not going to be spending very much power on keeping these caps charged up, so in practice, 4GB and 8GB capacities would end up having quite similar power draw. Probably well within the variability between two GPUs/graphics boards.

I thought I read some time ago AMD or another manufacturer confirmed there is something like 15W with each 4GB on GDDR5
Click to expand...
Yes, a full set of GDDR devices can draw quite a bit of power, because of the fast connection between memory devices and GPU. But that's separate from the DRAM array itself.

ImSpartacus said:
Or are you saying it's less than twice the power from the chips?
Click to expand...
Yes, that seems to be what he says. Instead of 200% power for 2 devices, it's like, 120-140%. Or that's the way I interpret it anyway. :)
 
ImSpartacus said:
Wait, clamshell mode generally consumes less power? Or are you saying it's less than twice the power from the chips? Very neat detail!
Click to expand...

I read that as each chip individually draws somewhat more than half of what it would if operating on its own. Since there are two of them in this mode the overall consumption is higher.
 
ToTTenTranz said:
As far as I can see, this is mostly a rumor/leak thread and it will continue to be one for another 9 days.
The latest leak comes from someone who took pictures of the card and claimed its power consumption was 110W.

I know this seems to be coming to bite in the ass of the users who were claiming Polaris was a lot less power efficient than Pascal and that Polaris was already a failure, but fret not: if the card turns out to be consuming 130, 140, or 150 or more watts, the thread title will surely be updated.
Click to expand...

Regardless, it seems to me that "<150W" is the only thing we know for sure at this point. That would be much less misleading than semi-random figures that change every week.
 
Silly season will probably continue for another week, until AMD lifts embargo.

Never have we had a situation where the premiere GPU about to be released only just matches the previous generation flagship. It's bizarre, and rather ridiculous as well.
 
Grall said:
Silly season will probably continue for another week, until AMD lifts embargo.

Never have we had a situation where the premiere GPU about to be released only just matches the previous generation flagship. It's bizarre, and rather ridiculous as well.
Click to expand...
The counterpoint is that Polaris is not the premier GPU, not relative to the top-end 28nm Nvidia or AMD GPUs.
In that regard, it's somewhat more like how we used to see pipe-cleaner or smaller GPUs used as the early movers for a new node. What might stand out now is how much AMD is hyping it.
 
3dilettante said:
The counterpoint is that Polaris is not the premier GPU, not relative to the top-end 28nm Nvidia or AMD GPUs.
In that regard, it's somewhat more like how we used to see pipe-cleaner or smaller GPUs used as the early movers for a new node. What might stand out now is how much AMD is hyping it.
Click to expand...
Yes, but to be fair, I don't think a pipe-cleaner has ever been as disruptive as Polaris 10 appears to be.
 
Apparently the score was with card at stock, the thread was pulled before overclocked results could be submitted.

Tie a strip bar, before he was driven to run 16.5.1 this old, I do not know under what's called a new drive, the graphics score is 5800 +, but the latest adaptation to tomorrow morning before driving to AMD China! Themselves and smoke it
Click to expand...

And something about agricultural products...

https://translate.google.co.in/tran....php?action=printable&tid=7217307&prev=search

https://translate.google.co.in/tran...bs.bigccq.cn/read.php?tid=9483158&prev=search
 
3dilettante said:
The counterpoint is that Polaris is not the premier GPU
Click to expand...
It's AMD's premier GPU, and it may be their premier GPU for the rest of the year for all we know. Their real flagship GPU isn't even a member of the same family and hasn't been showcased. In previous gens where middle-class or even low-end chips came first to market we knew what to expect from the flagship because it had already been announced previously.

Now there's going to be nothing, and probably for a long time.
 
Grall said:
It's AMD's premier GPU, and it may be their premier GPU for the rest of the year for all we know. Their real flagship GPU isn't even a member of the same family and hasn't been showcased. In previous gens where middle-class or even low-end chips came first to market we knew what to expect from the flagship because it had already been announced previously.

Now there's going to be nothing, and probably for a long time.
Click to expand...
How are you defining premier?
The numbers given so far in leaks and by AMD indicate it remains below Fiji.
It will be the newest GPU product on a new node, but it hasn't always been the out of the ordinary for that not to be the new high-end.
 
Status
Not open for further replies.

Similar threads

hoom
AMD: Zen 4, Speculation, Rumours and Discussion
8 9 10
Replies
199
Views
39K
MfA
M
Rys
Reopening Architecture and Products + rules and guidelines changes
3 4 5
Replies
90
Views
16K
Shifty Geezer
Shifty Geezer
D
AMD: Zen 2 (Ryzen/Threadripper 3000?, Epyc 8000?) Speculation, Rumours and Discussion
16 17 18
Replies
340
Views
153K
PSman1700
P
Shortbread
Next-Generation NVMe SSD and I/O Technology [PC, PS5, XBSX|S]
105 106 107
Replies
2K
Views
217K
Karamazov
K
L
A rambling history of AMD and nVidia's rise to the top
2
Replies
20
Views
6K
Leovinus
L
Back
Top