Urquan
09-13-2011, 12:31 AM
(09-12-2011, 10:54 PM)naoan Wrote: [ -> ]also your build won't launch at all with AMD Phenom II, thought you might want to know...
I can confirm this. Last one I had working was 3.0-71 SSE3.
09-13-2011, 12:31 AM
09-13-2011, 02:49 AM
Huh. That is good information, actually. The 3.0-71 SSE build was built with /arch:SSE3. The other 3.0-71 builds were built with /Qx and the appropriate SSE level. After that point, I built with /arch:SSE3 /QaXSSE4.2.
According to the intel compiler documentation
I suppose I'll switch to making plain /arch:SSE3 builds in addition to a SSE4.2 build then. In practice, I actually don't see much difference in performance between the two builds on my 2600k though. The compiler generated SSE4.1 and 4.2 instructions don't seem to offer real speedup.
Anyway, I am making a plain SSE3 build of that branch right now. Will throw it in the OP once it is done. Please confirm for me that it works as intended on a Phenom II. The only difference in my compilation is that I turned off /QaX.
According to the intel compiler documentation
intel docs Wrote:If you specify both the -ax and -m options (Linux) or the /Qax and /arch options (Windows), the baseline code will execute on non-Intel processors compatible with the processor type specified by the -m or /arch option.It should be making a separate codepath for SSE4.2. I certainly don't doubt either of you though, so it appears that something about it doesn't play nice with AMD processors.
The -ax and /Qax options tell the compiler to find opportunities to generate separate versions of functions that take advantage of features of the specified Intel® processor.
I suppose I'll switch to making plain /arch:SSE3 builds in addition to a SSE4.2 build then. In practice, I actually don't see much difference in performance between the two builds on my 2600k though. The compiler generated SSE4.1 and 4.2 instructions don't seem to offer real speedup.
Anyway, I am making a plain SSE3 build of that branch right now. Will throw it in the OP once it is done. Please confirm for me that it works as intended on a Phenom II. The only difference in my compilation is that I turned off /QaX.
Urquan
09-13-2011, 07:53 AM
(09-13-2011, 02:49 AM)FilthyMonkey Wrote: [ -> ]Huh. That is good information, actually. The 3.0-71 SSE build was built with /arch:SSE3. The other 3.0-71 builds were built with /Qx and the appropriate SSE level. After that point, I built with /arch:SSE3 /QaXSSE4.2.
According to the intel compiler documentation
intel docs Wrote:If you specify both the -ax and -m options (Linux) or the /Qax and /arch options (Windows), the baseline code will execute on non-Intel processors compatible with the processor type specified by the -m or /arch option.It should be making a separate codepath for SSE4.2. I certainly don't doubt either of you though, so it appears that something about it doesn't play nice with AMD processors.
The -ax and /Qax options tell the compiler to find opportunities to generate separate versions of functions that take advantage of features of the specified Intel® processor.
I suppose I'll switch to making plain /arch:SSE3 builds in addition to a SSE4.2 build then. In practice, I actually don't see much difference in performance between the two builds on my 2600k though. The compiler generated SSE4.1 and 4.2 instructions don't seem to offer real speedup.
Anyway, I am making a plain SSE3 build of that branch right now. Will throw it in the OP once it is done. Please confirm for me that it works as intended on a Phenom II. The only difference in my compilation is that I turned off /QaX.
The SSE3 build is working on my Phenom II.
09-13-2011, 02:04 PM
Are there 32bit versions of these builds? Thanks.
09-13-2011, 03:13 PM
No, and I am unlikely to make them. There simply isn't enough demand. There are other guys who fill that gap well enough too.
I mainly made the original build to see if the git head would resolve some crashing issues I was having in xenoblade from somebody else's older build. It turns out that it didn't help, but switching from xaudio to direct sound stopped the frequent crashing. I posted the builds on a forum I frequent, and here as a courtesy.
Really, I don't plan on making dolphin builds in the long term. I don't mind making builds for a while longer, but it is unlikely that you'll see the same sort of dedication from me as some of the other build producers here. I spend enough time compiling code in my professional life as it is. Consider my builds more of a stopgap solution.
I mainly made the original build to see if the git head would resolve some crashing issues I was having in xenoblade from somebody else's older build. It turns out that it didn't help, but switching from xaudio to direct sound stopped the frequent crashing. I posted the builds on a forum I frequent, and here as a courtesy.
Really, I don't plan on making dolphin builds in the long term. I don't mind making builds for a while longer, but it is unlikely that you'll see the same sort of dedication from me as some of the other build producers here. I spend enough time compiling code in my professional life as it is. Consider my builds more of a stopgap solution.
09-16-2011, 08:00 PM
How much of a difference does SSE3 make for AMD processors? I'd try it myself of course, but I'm at work and I'm impatient. :p
09-16-2011, 08:09 PM
yeah, it works fine now, thanks filthy.
@seek
not sure if it's because sse3 or newer revision but the 3.0-94 sse3 build is about 2-3 fps better than 7714M Xtreme2damax build.
@seek
not sure if it's because sse3 or newer revision but the 3.0-94 sse3 build is about 2-3 fps better than 7714M Xtreme2damax build.
09-17-2011, 03:48 PM
Hey, your builds produce bugs in Mario Golf (and possibly others). See here for some details:
http://code.google.com/p/dolphin-emu/issues/detail?id=3334#c41
http://code.google.com/p/dolphin-emu/issues/detail?id=3334#c41
09-17-2011, 06:05 PM
I just tested it myself. I get identical graphics in all of the builds. I took screenshots of the actual window, so the version numbers could be seen. All are using the dx9 graphics plugin. The putting grid problem can be seen in all of them.
my new-shadercache-uids build
ExtremeDude2's build labeled as new-shadercache-uids
my 3.0-96 build
ExtremeDude2's build labeled as 3.0-95
I say 'labeled as' because ExtremeDude2 builds seem to show strange revision numbers in the window title. His 3.0-95 build has 3.0-112 as the title.
An interesting note is the window title of the screenshot in the report you linked, purposing that the screenshot is showing bugs from the new-shadercache-uid branch. It is a screenshot from ExtremeDude2's 3.0-95 head build, as I don't think anything else shows up as 3.0-112 in the window title. This is the screenshot linked in that report. It is possible that the person making it linked the wrong screenshot.
The person mentioned that the problems went away when he switched to ExtremeDude2's branch build. I do not think that his 'branch build' (taken from this post ) is actually a branch build. It doesn't have 'new-shadercache-uids' in the window title as mine does, and if you check the about screen, it says that it is from "branch: master". Mine says "branch: new-shadercache-uids". That isn't something that I filled in myself.
my new-shadercache-uids build
ExtremeDude2's build labeled as new-shadercache-uids
my 3.0-96 build
ExtremeDude2's build labeled as 3.0-95
I say 'labeled as' because ExtremeDude2 builds seem to show strange revision numbers in the window title. His 3.0-95 build has 3.0-112 as the title.
An interesting note is the window title of the screenshot in the report you linked, purposing that the screenshot is showing bugs from the new-shadercache-uid branch. It is a screenshot from ExtremeDude2's 3.0-95 head build, as I don't think anything else shows up as 3.0-112 in the window title. This is the screenshot linked in that report. It is possible that the person making it linked the wrong screenshot.
The person mentioned that the problems went away when he switched to ExtremeDude2's branch build. I do not think that his 'branch build' (taken from this post ) is actually a branch build. It doesn't have 'new-shadercache-uids' in the window title as mine does, and if you check the about screen, it says that it is from "branch: master". Mine says "branch: new-shadercache-uids". That isn't something that I filled in myself.
09-18-2011, 12:08 AM
Uhm, could you please modify Source/Core/Common/Version.cpp (or wherever it is located, should be somewhere around there) to change the version string to sth like 3.0-94-ICC?
Just change
to something like
I appreciate the effort you're putting into this, but apparently it is causing lots of issues... for example http://code.google.com/p/dolphin-emu/issues/detail?id=4852 and the one that was mentioned before.
EDIT: Oh well, seems like that particular issue wasn't caused by ICC, but it'd be nice if you modified the version info of your builds like I said anyway.
Just change
Code:
const char *scm_rev_str = "Dolphin "
#if !SCM_IS_MASTER
"[" SCM_BRANCH_STR "] "
#endif
BUILD_TYPE_STR SCM_DESC_STR;Code:
const char *scm_rev_str = "Dolphin "
#if !SCM_IS_MASTER
"[" SCM_BRANCH_STR "] "
#endif
BUILD_TYPE_STR SCM_DESC_STR "-ICC";I appreciate the effort you're putting into this, but apparently it is causing lots of issues... for example http://code.google.com/p/dolphin-emu/issues/detail?id=4852 and the one that was mentioned before.
EDIT: Oh well, seems like that particular issue wasn't caused by ICC, but it'd be nice if you modified the version info of your builds like I said anyway.