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 1.93 seconds.
Search: Posts Made By: OpenGL guy
Forum: 3D Architectures & Chips 17-Apr-2014, 01:25
Replies: 1,639
Views: 230,605
Posted By OpenGL guy
GM107 has 1/32nd rate, not sure where you are...

GM107 has 1/32nd rate, not sure where you are seeing 1/128.
Forum: 3D Hardware, Software & Output Devices 02-Apr-2014, 16:15
Replies: 436
Views: 46,670
Posted By OpenGL guy
Your photos are still 3:2, are they not?

Your photos are still 3:2, are they not?
Forum: 3D Hardware, Software & Output Devices 02-Apr-2014, 15:21
Replies: 436
Views: 46,670
Posted By OpenGL guy
Photos are still 3:2. You can always view 16:9 at...

Photos are still 3:2. You can always view 16:9 at full resolution on 3:2 with black bars and top and bottom, not so the other way around. For coding, I prefer as much real estate as possible.
Forum: 3D Architectures & Chips 24-Feb-2014, 19:28
Replies: 1,639
Views: 230,605
Posted By OpenGL guy
Actually, Kepler did until a recent driver...

Actually, Kepler did until a recent driver release.
Forum: 3D Architectures & Chips 19-Feb-2014, 15:34
Replies: 1,639
Views: 230,605
Posted By OpenGL guy
The Radeon result is bogus. Upcoming drivers will...

The Radeon result is bogus. Upcoming drivers will "address" this issue.
Forum: 3D Architectures & Chips 17-Feb-2014, 17:07
Replies: 1,639
Views: 230,605
Posted By OpenGL guy
Counting SFUs where: In GFLOPS or reported number...

Counting SFUs where: In GFLOPS or reported number of ALUs?

The answer is neither. There are 64 ALUs per GCN CU. (To be precise, there are 16 ALUs for each of the 4 SIMDs within a CU.) Each ALU is...
Forum: 3D Architectures & Chips 15-Jan-2014, 22:11
Replies: 1,372
Views: 183,028
Posted By OpenGL guy
It's not as simple as that as it requires...

It's not as simple as that as it requires changing the VBIOS behavior at POST when it doesn't know what flavor of OS you will load later.

Yes, write-combine buffers were added to the North Bridge...
Forum: 3D Architectures & Chips 15-Jan-2014, 17:45
Replies: 1,372
Views: 183,028
Posted By OpenGL guy
We have to support our customers and that means...

We have to support our customers and that means people upgrading machines with older OSes.

No, not possible. What would you do if multiple applications were trying to access different 256MB...
Forum: 3D Architectures & Chips 15-Jan-2014, 17:05
Replies: 1,372
Views: 183,028
Posted By OpenGL guy
Thanks to 32-bit OSes, only a portion of the...

Thanks to 32-bit OSes, only a portion of the graphics card's memory is exposed to the CPU. For AMD parts, this is around 256MB of the total video memory. This means that the CPU cannot directly...
Forum: 3D Architectures & Chips 03-Dec-2013, 19:41
Replies: 2,395
Views: 297,201
Posted By OpenGL guy
Inefficient compared to what? FAHBench...

Inefficient compared to what?

FAHBench results in ns/day:
R9-290X: 48.3 explicit, 178.4 implicit
Intel i7-4770K: 4.2 explicit, 4.6 implicit
Forum: 3D Architectures & Chips 25-Oct-2013, 20:19
Replies: 2,395
Views: 297,201
Posted By OpenGL guy
290X has 1:8 DP.

290X has 1:8 DP.
Forum: 3D Architectures & Chips 11-Oct-2013, 19:33
Replies: 2,395
Views: 297,201
Posted By OpenGL guy
Tahiti has two shader engines,each with two...

Tahiti has two shader engines,each with two shader arrays. See http://forum.beyond3d.com/showthread.php?p=1785875#post1785875.
Forum: 3D Architectures & Chips 07-Oct-2013, 19:58
Replies: 1,372
Views: 183,028
Posted By OpenGL guy
Really? This argument fails: - The 8GB of...

Really?

This argument fails:
- The 8GB of slower system memory you would have bought is not free
- The discrete card adds more cost since you need to pair it with a CPU as well
- The discrete...
Forum: 3D Architectures & Chips 07-Oct-2013, 16:54
Replies: 1,372
Views: 183,028
Posted By OpenGL guy
Wow, you guys are really grasping at straws here....

Wow, you guys are really grasping at straws here. APUs are cheaper for the following reasons:
- No reason to buy a discrete graphics card with its own memory
- Shares power with the CPU
- Shares...
Forum: 3D Architectures & Chips 06-Oct-2013, 08:16
Replies: 1,372
Views: 183,028
Posted By OpenGL guy
Of course it does. You don't need 8 GB to run...

Of course it does. You don't need 8 GB to run your machine, for one thing. Secondly, does the discrete card come with a CPU? Didn't think so.
Forum: 3D Architectures & Chips 05-Oct-2013, 20:21
Replies: 1,372
Views: 183,028
Posted By OpenGL guy
You're comparing 8GB of high speed RAM to a GPU...

You're comparing 8GB of high speed RAM to a GPU board with 1GB of memory. Do you expect to run your whole system on 1GB of memory including the part the GPU uses?
Forum: 3D Architectures & Chips 29-Sep-2013, 17:36
Replies: 2,395
Views: 297,201
Posted By OpenGL guy
What was the source for the chart? I ask because...

What was the source for the chart? I ask because the PCIe gen 2 had a lot of command overhead (like 20% of the bandwidth or something) and that was reduced with PCIe gen 3. In fact, I believe a...
Forum: 3D Architectures & Chips 29-Sep-2013, 16:57
Replies: 2,395
Views: 297,201
Posted By OpenGL guy
I've never seen anything beyond about 12.8 GB/s...

I've never seen anything beyond about 12.8 GB/s for a single directional transfer on PCIe gen 3.0 and 16 lanes.
Forum: 3D Architectures & Chips 23-Sep-2013, 13:53
Replies: 2,395
Views: 297,201
Posted By OpenGL guy
Channel stride is 256 bytes. This is documented...

Channel stride is 256 bytes. This is documented in the Stream SDK Programmer's Guide.
Forum: 3D Architectures & Chips 19-Sep-2013, 17:41
Replies: 2,395
Views: 297,201
Posted By OpenGL guy
I only used the terminology you used above. ...

I only used the terminology you used above. Pitcairn and Tahiti have 4 shader arrays (GCN parlance).

Tahiti's organization of CUs within each shader array is simple because there are 8 CUs per...
Forum: 3D Architectures & Chips 19-Sep-2013, 17:10
Replies: 2,395
Views: 297,201
Posted By OpenGL guy
Pitcairn has 4 CU groups as has Tahiti.

Pitcairn has 4 CU groups as has Tahiti.
Forum: 3D Architectures & Chips 28-Aug-2013, 19:39
Replies: 11
Views: 2,584
Posted By OpenGL guy
Security for one. It's also helpful for OpenCL. ...

Security for one. It's also helpful for OpenCL. I am not going to go into further details.
Forum: 3D Architectures & Chips 28-Aug-2013, 16:57
Replies: 11
Views: 2,584
Posted By OpenGL guy
No, if the GPU triggered a page fault, there...

No, if the GPU triggered a page fault, there would have to be OS support to handle it, which there isn't.

Virtualization has nothing to do with swap, it just means that the addresses used to...
Forum: 3D Architectures & Chips 28-Aug-2013, 16:53
Replies: 11
Views: 2,584
Posted By OpenGL guy
Cache coherent with what? The CPU? No, the GPU...

Cache coherent with what? The CPU? No, the GPU does not share the CPU's page tables.
Forum: 3D Architectures & Chips 26-Aug-2013, 06:54
Replies: 11
Views: 2,584
Posted By OpenGL guy
No.

No.
Showing results 1 to 25 of 500

 
Forum Jump

All times are GMT +1. The time now is 20:14.


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