#4583329 Posted on: 04/24/2013 05:08 PM

---

Nice

#4583332 Posted on: 04/24/2013 05:09 PM

---

Holy crap, I was just about to go to bed, thinking I missed out on it... and it appears! Sweet. Let's see if those frame latency fixes have much of an affect on my 5970/5850 combo.

Edit: Latter URL am broke. One thing I can see quickly before heading to bed, most files have the date of 17/4, if that means anything.

#4583339 Posted on: 04/24/2013 05:19 PM

---

You have to wait a little bit for the release note, amd needs to update the web page first.

#4583340 Posted on: 04/24/2013 05:19 PM

---

desktop link is broke, mobility link works.

backwards from ususal.

Usually the mobility people get screwed.

lol

#4583341 Posted on: 04/24/2013 05:20 PM

---

Is this newer or older than 13.4? Atleast driver version was lower.

#4583342 Posted on: 04/24/2013 05:20 PM

---

Nice find OP

#4583343 Posted on: 04/24/2013 05:20 PM

---

Holy crap, I was just about to go to bed, thinking I missed out on it... and it appears! Sweet. Let's see if those frame latency fixes have much of an affect on my 5970/5850 combo.

Edit: Latter URL am broke. One thing I can see quickly before heading to bed, most files have the date of 17/4, if that means anything.

The computer date and time stamps for files are stored in UTC (GMT). The date and timestamps you see in the files is offset from this according to the regional time settings on the computer. Since you, like myself, are in Australia, the date and timestamp is (currently) between +8 and +10 UTC (you +10 UTC, me +9.5 UTC). This means that even if the driver had a date/timestamp of 16 April in America, it could well be 17 April here (since we are up to 18 hours ahead of US Pacific time). The date in the inf is of course, fixed.

#4583347 Posted on: 04/24/2013 05:24 PM

---

Components version:

CCC: 2013.0416.1036.17145
D3D: 9.14.10.0968
OGL: 6.14.10.12198
OCL: 10.0.1124.2 (Driver Integrated)

#4583348 Posted on: 04/24/2013 05:25 PM

---

Mirror site:
http://www.computerbase.de/downloads/treiber/grafik/amd-catalyst-vorab-treiber/

#4583352 Posted on: 04/24/2013 05:26 PM

---

Components version:

CCC: 2013.0416.1036.17145
D3D: 9.14.10.0968
OGL: 6.14.10.12198
OCL: 10.0.1124.2 (Driver Integrated)

Anything are older then catalyst 13.4 whql, expect for the ccc

#4583353 Posted on: 04/24/2013 05:29 PM

---

I have installed these on my 6870 1gb in W7 64, no problems.

#4583355 Posted on: 04/24/2013 05:33 PM

---

These drivers were quite unstable when sweclockers.com did their 7990 review...
The frame times are better in many games but stability comes first.

#4583356 Posted on: 04/24/2013 05:33 PM

---

D3D: Appear to be older than 13.4 WHQL

#4583357 Posted on: 04/24/2013 05:34 PM

---

Components version:

CCC: 2013.0416.1036.17145
D3D: 9.14.10.0968
OGL: 6.14.10.12198
OCL: 10.0.1124.2 (Driver Integrated)

These are actually older than the Catalyst 13.4 released driver then! (with newer components).

Ironically, I copied over the CCC etc from the previous 16 April driver, which is version 2013.0416.1149.19347, so that is newer too.

AVT and WMVDecoder are also slightly older builds in this driver compared to the 16 April beta, by a couple of hours. The file sizes are also different for the 16 April beta, they are not the same!

It appears that this beta is merely an official release, under a different Catalyst label, of a slightly older build of the 16 April beta. I would recommend sticking with using Catalyst 13.4 with the 16 April apps

(except for OpenCL which is now incorporated)

.

#4583359 Posted on: 04/24/2013 05:36 PM

---

Will stick with the 13.4's if these are older then.

#4583362 Posted on: 04/24/2013 05:38 PM

---

Per Thracks, the 13.4s are just 13.3bet3 with the 7990 ASIC added. No other changes.

I wouldnt go by the date.

If these are later than 13.3beta3, then they are newer.

#4583370 Posted on: 04/24/2013 05:47 PM

---

The 13.4's are different to the 13.3beta 3. I don't care if AMD claims that the 7990 ASIC is the only difference, because like I said in the 13.4 thread, most changes go unreported. If the only reported changes were what AMD said there were, then drivers haven't really changed much in the last several years.

Go by the build number. The build number for the Catalyst 13.4 are higher. The lower build numbers mean that although it may be compiled later, the compilation is done with earlier code. AMD do make minor changes to the drivers without changing the build numbers, but the 13.4 driver is definitely more up to date.



The REAL FACT is that the 13.5b2's are the same as the 13.3b3 with the addition of the 7990 ASIC (since 13.5b2 is a HD7990 beta driver), and updated apps, and the OpenCL dll's installed as part of the driver. I am talking the actual driver here, not the whole package that includes the apps.

The 13.4 drivers are absolutely and most definitely newer! (except for the apps). The apps are separate to the driver, they package the latest apps available at the time of packaging.

The only way to get the latest combination is to use the 13.4 WHQL driver, and the 13.3b3 16 April apps (except for OpenCL). The 13.4 installer will install the new apps without hassle.

Don't be fooled by the '13.5 beta 2'. That is a packaging designation, NOT a driver version.

#4583392 Posted on: 04/24/2013 06:10 PM

---

i disagree.

the 13.5b2 are PROTOTYPE drivers, meaning they have special code in them.

I can GUARANTEE if you are using xfire, you want them over 13.4

considering 13.4 and 13.5b2 have added both 7790, and 7990 ASICS, and 13.5b2 has prototype code, regraless of the numbering and compile dates scheme, i believe 12.5b2 are newer.

im sure thracks will comment eventually

#4583401 Posted on: 04/24/2013 06:20 PM

---

so what the best ccc with drivers 13.4?

#4583406 Posted on: 04/24/2013 06:28 PM

---

I can GUARANTEE if you are using xfire, you want them over 13.4
While I sincerely hope you're correct (and I'm not saying you're wrong!), has anyone actually verified, for a fact, that the 13.5b2 drivers include additional fixes and profiles for Crossfire that are not present in 13.4 WHQL?

#4583408 Posted on: 04/24/2013 06:29 PM

---

haaa!!!!

hmm Odd stuff, Direct3D and OGL are older than 13.4..!!!

need some benchmark differences here...!!! 13.4 vs 13.5 B2..

#4583415 Posted on: 04/24/2013 06:31 PM

---

If that is the case, it does kind of make sense, since the HD7990 is effectively Crossfired, just in the same card! That is probably why the rest of the driver is older, they started with that code and added the special HD7990 code to it. For all non-crossfired cards, the 13.4's are still the better option.

Of course, once the testing is complete the codelines will be amalgamated.

#4583456 Posted on: 04/24/2013 06:59 PM

---

13.5b2 + 13.4c1 app profiles 24apr

it's the only way for ultra smooth gaming for now...!!!

#4583472 Posted on: 04/24/2013 07:17 PM

---

13.5b2 + 13.4c1 app profiles 24apr

it's the only way for ultra smooth gaming for now...!!!
You say that every single time a driver is released? LOL

#4583473 Posted on: 04/24/2013 07:17 PM

---

If that is the case, it does kind of make sense, since the HD7990 is effectively Crossfired, just in the same card! That is probably why the rest of the driver is older, they started with that code and added the special HD7990 code to it. For all non-crossfired cards, the 13.4's are still the better option.

Of course, once the testing is complete the codelines will be amalgamated.

i guess that would depend on weather or not the prototype code has any positive effect for single cards.

from what ive read, its a frame metering technique to stablize the frame return latency, in which case you would need 2 simultaneous render pipleines to make it effective.

This would rule out any benefit for single cards.