Welcome, Unregistered.

If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below.

Showing results 1 to 25 of 500
Search took 4.22 seconds.
Search: Posts Made By: 3dilettante
Forum: 3D Architectures & Chips 29-Sep-2014, 19:03
Replies: 2,696
Views: 394,968
Posted By 3dilettante
Is (was/will) the lack of time-travel capability...

Is (was/will) the lack of time-travel capability a capital offense these (those?) days?
Forum: 3D Architectures & Chips 25-Sep-2014, 01:31
Replies: 169
Views: 34,220
Posted By 3dilettante
If it's first-gen HBM, getting to 4 GB would...

If it's first-gen HBM, getting to 4 GB would require four stacks at 128 GB/s bandwidth (some fiddling may come from speed grade) each.

A 2-stack design would get 256 GB/s in bandwidth, which is...
Forum: 3D Architectures & Chips 22-Sep-2014, 21:28
Replies: 11
Views: 1,459
Posted By 3dilettante
Because the metal layer pitch didn't improve for...

Because the metal layer pitch didn't improve for the foundries? It wouldn't be the first time in the last several nodes that a company took its foot off the gas in terms of wire pitch.
Forum: 3D Architectures & Chips 22-Sep-2014, 20:15
Replies: 2,696
Views: 394,968
Posted By 3dilettante
Since this is being framed as a question about...

Since this is being framed as a question about the general future of AMD and the tech in question is not directly tied to a GPU or GPU family, there is a doom and gloom thread for AMD:
...
Forum: 3D Architectures & Chips 22-Sep-2014, 17:21
Replies: 11
Views: 1,459
Posted By 3dilettante
Process node numbers have been made up for years...

Process node numbers have been made up for years now by everybody.
There's no single feature that physically matches the numbers used for node names, just some sort of "effective" improvement.
Forum: 3D Architectures & Chips 20-Sep-2014, 21:30
Replies: 162
Views: 19,635
Posted By 3dilettante
A very wide bus, very fast bus, or some...

A very wide bus, very fast bus, or some combination of the two is quite possible with upcoming levels of package/interposer integration.
The GPU makers seem to be opting for a very wide bus with...
Forum: 3D Architectures & Chips 20-Sep-2014, 20:59
Replies: 2,390
Views: 375,641
Posted By 3dilettante
THG claims its compute load had a sustained...

THG claims its compute load had a sustained average draw above TDP, but I'm uncertain this isn't a case of playing with a new toy in the wrong way.


You can't do much to TDP, because it's just a...
Forum: Console Technology 20-Sep-2014, 04:58
Replies: 51
Views: 3,950
Posted By 3dilettante
I would say the general lack of peer status for...

I would say the general lack of peer status for the slave GPU would be one of the primary reasons why it isn't considered a CPU. Its compute units might meet a general definition of being a core,...
Forum: 3D Architectures & Chips 19-Sep-2014, 23:01
Replies: 2,390
Views: 375,641
Posted By 3dilettante
It's squiggly lines with no particular sign of...

It's squiggly lines with no particular sign of context or understanding, and it makes comparisons to other architectures that it has neglected to profile and above all else failed to log something as...
Forum: 3D Architectures & Chips 19-Sep-2014, 22:31
Replies: 2,390
Views: 375,641
Posted By 3dilettante
You don't need a high-speed oscilloscope if the...

You don't need a high-speed oscilloscope if the power draw averaged over a few milliseconds is above TDP.
It's a specification that adjusts for the way a cooler's physical bulk will smear together...
Forum: 3D Architectures & Chips 19-Sep-2014, 22:11
Replies: 2,390
Views: 375,641
Posted By 3dilettante
There's an instantaneous power cost to wakeup,...

There's an instantaneous power cost to wakeup, especially for power gating. Power gates themselves have a power cost when they switch, especially since they need to be physically larger than most...
Forum: 3D Architectures & Chips 19-Sep-2014, 20:11
Replies: 2,390
Views: 375,641
Posted By 3dilettante
Possibly things like high utilization where a lot...

Possibly things like high utilization where a lot of SIMD units switch at the same time, current inrush from a ton of clock-gated units waking up, a confluence of high ALU activity, high memory bank...
Forum: 3D Architectures & Chips 19-Sep-2014, 19:51
Replies: 2,390
Views: 375,641
Posted By 3dilettante
If the chip doesn't maintain above TDP power draw...

If the chip doesn't maintain above TDP power draw for periods that measure more than a few milliseconds, if none of the transient spikes exceed the maximum power rating (not the same thing as TDP),...
Forum: 3D Architectures & Chips 19-Sep-2014, 17:32
Replies: 2,390
Views: 375,641
Posted By 3dilettante
Any device in this class can, because it's not a...

Any device in this class can, because it's not a challenge for these chips to physically draw way more power than what they are rated for. It's not particularly challenging for them to draw enough...
Forum: 3D Architectures & Chips 19-Sep-2014, 05:13
Replies: 50
Views: 6,769
Posted By 3dilettante
Intel's PixelSync, or rather the pixel shader...

Intel's PixelSync, or rather the pixel shader ordering that goes into it, is something I've not seen mentioned for competing GPUs.
One of the first things noted about DX12 was the introduction of a...
Forum: 3D Architectures & Chips 10-Sep-2014, 09:28
Replies: 50
Views: 6,769
Posted By 3dilettante
According to the presentation it can, which is...

According to the presentation it can, which is why it appears to be better.
Forum: 3D Architectures & Chips 10-Sep-2014, 08:08
Replies: 50
Views: 6,769
Posted By 3dilettante
It's a little sparse on detail, but the GPU L3...

It's a little sparse on detail, but the GPU L3 and its system interface leverage virtual memory support and IO virtualization to provide shared virtual memory, which seems similar at a high level to...
Forum: 3D Architectures & Chips 08-Sep-2014, 21:31
Replies: 2,390
Views: 375,641
Posted By 3dilettante
The thing about virtualization is that it's...

The thing about virtualization is that it's implemented in hardware that is non-negotiable as far as x86 CPU cores go and it's not a significant increase in hardware cost. It can be fused on or off...
Forum: 3D Architectures & Chips 08-Sep-2014, 19:59
Replies: 2,390
Views: 375,641
Posted By 3dilettante
It's not impossible to have separate silicon for...

It's not impossible to have separate silicon for bifurcation based on a single metric, in this case DP throughput.
However, if everything else is equal, you will stand a good chance of paying...
Forum: Console Technology 05-Sep-2014, 17:28
Replies: 820
Views: 93,944
Posted By 3dilettante
SRAM doesn't have to be on the same die either,...

SRAM doesn't have to be on the same die either, but a second custom die and an MCM adds complexity and cost to the whole project.
The DF interview indciated that Microsoft wanted a single-chip...
Forum: 3D Architectures & Chips 04-Sep-2014, 16:58
Replies: 162
Views: 19,635
Posted By 3dilettante
The data bus is DDR. It is clocked at twice the...

The data bus is DDR. It is clocked at twice the command clock.
Forum: Console Technology 03-Sep-2014, 19:22
Replies: 131
Views: 15,650
Posted By 3dilettante
The cited image is actually a very interesting...

The cited image is actually a very interesting illustration of the concept by providing a mapping of a GPU's throughput (or rather, one SIMD in the excerpt) where color along the vertical axis...
Forum: 3D & Semiconductor Industry 03-Sep-2014, 18:47
Replies: 9
Views: 935
Posted By 3dilettante
I read that part of the quote, but I was curious...

I read that part of the quote, but I was curious as to whether there was some level of feedback such that the quality of both sides improves as a result.
Forum: 3D & Semiconductor Industry 03-Sep-2014, 18:24
Replies: 9
Views: 935
Posted By 3dilettante
The device description matches Cape Verde. ...

The device description matches Cape Verde.

Is Matrox going to run with the drivers AMD puts out, or will there be some level of codevelopment between their software and AMD's driver?
Alternately,...
Forum: 3D Architectures & Chips 03-Sep-2014, 18:14
Replies: 2,390
Views: 375,641
Posted By 3dilettante
Why not do both? That seems like a win-win.

Why not do both?
That seems like a win-win.
Showing results 1 to 25 of 500

 
Forum Jump

All times are GMT +1. The time now is 16:51.


Powered by vBulletin® Version 3.8.6
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.