huh? Are you saying I'm going to have trouble with Pd itself on the Athon64? I've never heard of using PiDiP without Pd.

I'm not so worried about compiling for 64 bit or anything. I just want to make sure the software will run well before I spend installation time and start building on hardware that won't support my needs.

-John

cdr wrote:

On Sun Jan 22, 2006 at 08:50:29PM -0600, John Harrison wrote:

...

I noticed that the effectv site mentions support only for Intel IA-32 architecture. Does this apply to PiDiP, then, as well? So...if I use PiDiP, for example, on an Athon machine, am I going to get inferior or unwanted performance?

if theres a version of PiDiP that doesnt require PD, it may work right on Athlon64..

...

-John

---

PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list

---

PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list

Le 22 Janvier 2006 22:24, cdr a écrit :

on x86_64 if your needs include running pure data you will want a 32bit chroot or multilib setup, its pretty simple and there are wiki pages explaining how to do this on debian and gentoo. ALSA and 3D hardware will still work..

For Debian I found this documentation: http://alioth.debian.org/docman/view.php/30192/21/debian-amd64-howto.html There's a section "Using an IA32 chroot to run 32bit applications". -- Marc

I guess what I was really wondering is if there are libraries (like MMX or whatever) which the PiDiP code uses and are available only for the Intel chips and not the AMD. I'm not super up on the nuts and bolts but I thought I had recalled that AMD and Intel had non-compatible multimedia extensions (as of something like 5 years ago...maybe things have changed) and I just wanted to make sure I'll get reasonable performance on the AMD, or if PiDiP is tailored to Intel-specific architecture. This is all because I saw that statement about IA-32 on effectv and it made me nervous.

I don't know if I am hard core enough to go 64 bit OS...unless somebody tells me the performance will skyrocket...but I've made a note of this Debian link.

Thanks for the help,

-John

Hans-Christoph Steiner wrote:

On Jan 22, 2006, at 11:13 PM, Marc Lavallée wrote:

...

Le 22 Janvier 2006 22:24, cdr a écrit :

...

on x86_64 if your needs include running pure data you will want a
32bit chroot or multilib setup, its pretty simple and there are wiki pages explaining how to do this on debian and gentoo. ALSA and 3D hardware will still work..

For Debian I found this documentation: http://alioth.debian.org/docman/view.php/30192/21/debian-amd64- howto.html There's a section "Using an IA32 chroot to run 32bit applications".

You can just run them in x86 mode and it should behave just like any
other x86 AFAIK. You only need this fancy chroot stuff if you want to run a 64-bit OS.

.hc

---

---

                        http://at.or.at/hans/

---

PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list

On Jan 23, 2006, at 8:52 AM, Thomas Sivertsen wrote:

John Harrison wrote:

...

and I just wanted to make sure I'll get reasonable performance on the AMD, or if PiDiP is tailored to Intel-specific architecture. This is all because I saw that statement about IA-32 on effectv and it made me nervous.

An Athlon64 should work since PiDiP is not effectv. Some effects are
ported from Effectv. Requirements for effectv have little to do with
PiDiP. Besides, most MMX and SSE stuff is supported on Athlons and
have for several years. Just because a few of the effects have been
ported does not mean PiDiP is a port of effectv to PD. PiDiP contains
a lot more than just some nifty effects.

Who started saying this thing about PiDiP being a port of effectv
anyway? it needs to stop, right now.

I first brought up effectv saying that effectv code was used in PiDiP,
therefore PiDiP must be released under the GNU GPL. I don't know where
the idea of pidip being an effectv port came from.

.hc

News is what people want to keep hidden and everything else is
publicity.

                  - Bill Moyers

Marc Lavallée skrev:

What's the problem about porting and extending software, or using code from some other software to kickstart a more ambitious project? That's a very common practice with free software. Look at the changelog.

That is not what i said. i was trying to get the point across that pidip is not effectv, and that system requirements for effectv has absolutely nothing to do with pidip. And i was pretty much saying, allthough not in so many words, the exact same thing as you are. it's one of the great benefits of open-source software. The problem was that someone thinks pidip is somehow only effectv, which it is not.

.thomas