All times are UTC-06:00




Post new topic  Reply to topic  [ 17 posts ] 
Author Message
 Post subject: OSW Uncovered...
PostPosted: Sat Sep 23, 2006 10:25 am 
Offline
Genesi

Joined: Fri Sep 24, 2004 1:39 am
Posts: 1422
The Cat is Out of the Bag!

R&B :D

_________________
http://bbrv.blogspot.com


Top
   
 Post subject:
PostPosted: Sat Sep 23, 2006 10:33 am 
Offline

Joined: Tue Aug 15, 2006 5:08 am
Posts: 50
Location: France
We want photo, configurations... price !!

We want to know evertything !!


:wink:

:D


Top
   
 Post subject:
PostPosted: Sat Sep 23, 2006 11:52 am 
Offline

Joined: Mon Nov 28, 2005 2:51 pm
Posts: 41
Location: Dallas
I'm glad to hear that the OSW has what sounds to be a big focus on next-generation console development and game development in general. It's probably too late to make changes to whatever release plans that are put in place for the OSW (I wish we all new more details about OSW's release), but I'd like to add a request anyway :D.

I would like the ability to purchase the individual parts separately (motherboard, processor, memory, etc.). I'm under the impression though that the motherboard and processor will not be sold separately and that the processor will not be upgradeable. The only reason I've bought only x86 computers in the past is because of the ability to change out, replace, and upgrade parts as I wanted (Windows used to be another reason, but that hasn't been the case in a while). I think bringing over that ability to get more involved with the hardware would bring a huge boost in the adoption of Power architecture.

Take a look at how many x86-related hardware sites there are on the internet... most of them don't come out and say x86-only though, it's just understood. Why is that? Wouldn't it be great if that wasn't the case? Imagine if on every weekly CPU Buying Guide there was a section for Power processors, and on every hardward site they included Power-based components as well. Linux is already starting to become included, with hardware-centric sites such as Anandtech including a Linux section and multiple Linux-centric gaming sites (like Happy Penguin,www.linux-gamers.net, and Linux Game Publishing) becoming more popular.

With these new technologies and developments it doesn't seem like it should be too far from happening:
-Genesi's Firmware, providing the ability for graphics cards to be useable on both x86 and Genesi's Power-based systems.
-QuickTransit Virtualization, being included on IBM's upcoming OpenPower and p5 systems in 2007 which will allow x86-compiled binaries to run on Power with as little as 20% performance decrease.
-Glibc PowerPC Optimizations, which will greatly increase the performance of memory-bound applications on Power architecture.

_________________
Joshua Purcell


Top
   
 Post subject:
PostPosted: Sat Sep 23, 2006 3:13 pm 
Offline
Site Admin

Joined: Fri Sep 24, 2004 1:39 am
Posts: 1589
Location: Austin, TX
Quote:
With these new technologies and developments it doesn't seem like it should be too far from happening:
-Genesi's Firmware, providing the ability for graphics cards to be useable on both x86 and Genesi's Power-based systems.
-QuickTransit Virtualization, being included on IBM's upcoming OpenPower and p5 systems in 2007 which will allow x86-compiled binaries to run on Power with as little as 20% performance decrease.
-Glibc PowerPC Optimizations, which will greatly increase the performance of memory-bound applications on Power architecture.
Remember http://www.freevec.org/ has optimizations for glibc (these will be ported to glibc proper soon) which far outweigh those on Gunnar's site. Not that he did a bad job, but he just cleaned up the crap; there's nothing like a bit of hand=vectorized code to speed things up though.

You're right the processor won't be upgradable on this version; we are trying to keep costs down. There are enough expansion ports for everything else though. We feel that the market is used to Apple boxes; upgrading the CPUs on the newer ones was all but impossible (nobody made them). The same is true of Pegasos; an upgradable board with only 1 available processor is not truly upgradable. We need to create a market for it before it can happen, otherwise it is a waste of everyone's time (including yours).

_________________
Matt Sealey


Top
   
 Post subject:
PostPosted: Sat Sep 23, 2006 3:29 pm 
Offline

Joined: Sun May 08, 2005 8:46 pm
Posts: 559
Location: Paris
Code:
The same is true of Pegasos; an upgradable board with only 1 available processor is not truly upgradable. We need to create a market for it before it can happen, otherwise it is a waste of everyone's time (including yours).
true, but the "upgradable cpu card" feature is still used as a marketing weapon on Genesi's webpages.
There may not be a market for it, but it still made a few disappointed users. Unless bbrv's offer finally comes true.

I don't think the OSW needs the same feature, soldered cpu(s) is just fine. Has the OSW price changed since the first announcements ? can we have an update on it ?
need to save...


Top
   
 Post subject:
PostPosted: Sat Sep 23, 2006 8:45 pm 
Offline

Joined: Wed Feb 22, 2006 1:59 am
Posts: 180
Location: Australia
very interesting, if it turns out it could be a real push for PPC as a platform. Coincidentally after the demise of e3 a crop of these cropped up including go3 in perth.

EDIT: Link
http://www.go3.com.au/


Top
   
 Post subject:
PostPosted: Sun Sep 24, 2006 2:00 am 
That's great !
I hope the OSW will be available soon.


Top
   
 Post subject:
PostPosted: Sun Sep 24, 2006 11:10 am 
Offline

Joined: Thu Mar 16, 2006 10:02 am
Posts: 32
Sorry guys i only see a cat in the bag...looks like a re- posted old schematic of the OSW.

(Trying to get the OSW specs out of the bag from BBRV or Neko mode=on)
So what do we have gathered in different threads?

:!:
Dual G5/PPC970MP@1,25GHz (maybe more)
"Southbridge function" by Broadcom HT-1000 and HT-2000
1 x PCIe x16 Slot
1 x PCIe x8 Slot
2? x GigE ports
"More than 2" SATA ports
USB Ports

Speculation:
XGI Graphics Card with fully supported 3D functions, and maybe video acceleration.

(Trying to get the OSW specs out of the bag from BBRV or Neko mode=off)

I once saw a price list indicating that the "sweet spot" price wise for the PPC970 was the one at 1,8GHz so i hope the OSW gets at least them inside.


Top
   
 Post subject:
PostPosted: Sun Sep 24, 2006 12:28 pm 
Offline
Genesi

Joined: Fri Sep 24, 2004 1:39 am
Posts: 1422
Hi Donar, here are a couple more tidbits...
Quote:
PROPOSAL: We would like to name _the Dual970 Developer Platform_ an Atari. We can target all next generation consoles and we can add open source developers and software to the publishing mix. Together, we can put the tweaks in place to make Atari distribution desirable and elemental. If anything, your publishing business will have a new and fresh source of content. As we work closely with IBM, we can get them to support our efforts into all three primary console developments. We can also provide a development environment for GBA, Nintendo DS and even the Retro market if you want to address it (how about start here with sources?).
That was one approach. We are still working on it...

Atari, Amiga, whatever with the same core in all three major consoles this workstation will be successful.

R&B :)

_________________
http://bbrv.blogspot.com


Top
   
 Post subject:
PostPosted: Mon Sep 25, 2006 3:06 am 
Offline

Joined: Wed Feb 22, 2006 1:59 am
Posts: 180
Location: Australia
tomato/tomaeto (german pronunciation) its all the same thing really. The key point in this case is brand building. Creating some good applications for the platform increases your user base which when happy increases brand awareness, which encourages more developers to create better applications. (Chicken and the egg scenario)

What would be the target market for the OSW anyway?


Top
   
 Post subject:
PostPosted: Mon Sep 25, 2006 3:07 am 
Offline
Site Admin

Joined: Fri Sep 24, 2004 1:39 am
Posts: 1589
Location: Austin, TX
Quote:
Atari, Amiga, whatever with the same core in all three major consoles this workstation will be successful.
It doesn't matter what we call it, if you want to develop for XBox, you buy an XBox development kit. You can't get your game certified for sale without one. The same for the Sony PS3 development kit. The same for the Wii development kit.

I don't think game console developers will bite if they need to buy a dual-core G5 workstation (which Microsoft did ship for a while as an interim development kit) AND the official kit for each supplier.

It's also true that buying the hardware does NOT mean you have a license to publish games on the console.

_________________
Matt Sealey


Top
   
 Post subject:
PostPosted: Mon Sep 25, 2006 4:10 am 
Offline
Genesi

Joined: Fri Sep 24, 2004 1:39 am
Posts: 1422
We are talking about something different here. You will see information on Power.org about this opportunity in the months ahead.

R&B

_________________
http://bbrv.blogspot.com


Top
   
 Post subject:
PostPosted: Mon Sep 25, 2006 9:20 pm 
Offline

Joined: Mon Nov 28, 2005 2:51 pm
Posts: 41
Location: Dallas
Quote:
You're right the processor won't be upgradable on this version; we are trying to keep costs down. There are enough expansion ports for everything else though. We feel that the market is used to Apple boxes; upgrading the CPUs on the newer ones was all but impossible (nobody made them). The same is true of Pegasos; an upgradable board with only 1 available processor is not truly upgradable. We need to create a market for it before it can happen, otherwise it is a waste of everyone's time (including yours).
If Genesi is the only company who would be releasing CPUs for this board, then the market still wouldn't be the same as if there were many different options to choose from. But it would at least be a start, and a future upgrade (like the recent 7447A upgrade for Pegasos2) may be easier. I guess there's always hope for an add-on Cell card :D.

_________________
Joshua Purcell


Top
   
 Post subject:
PostPosted: Tue Sep 26, 2006 8:20 pm 
Offline

Joined: Mon Nov 28, 2005 2:51 pm
Posts: 41
Location: Dallas
Quote:
Quote:
-Glibc PowerPC Optimizations, which will greatly increase the performance of memory-bound applications on Power architecture.
Remember http://www.freevec.org/ has optimizations for glibc (these will be ported to glibc proper soon) which far outweigh those on Gunnar's site. Not that he did a bad job, but he just cleaned up the crap; there's nothing like a bit of hand=vectorized code to speed things up though.
I not sure how these two projects fit in with each other. From what it sounds like, both are attempting to do the same thing (libfreevec is just further along). I did have a question about this project though:

-Does the end goal of the freevec project include getting these optimizations into glibc?

-Would it be beneficial to the freevec project to start a discussion over at Power.org about the work being done? I bring this up because there may be others there who would be willing to help in some way, and the only thing I've seen that is somewhat related to this is Gunnar's posts about his glib optimizations. There was an interesting discussion a month or two ago about this, and freevec was only mentioned in passing.

-I know there will always be multiple paths to the same goal (multiple Linux distributions for example), but what specifically are the difference between the work Gunnar has done and the freevec project? If they are the same thing working towards the same goal, then has there ever been any communication between the two projects?

I'm interested in this because all the talk I've heard about Power-optimized Linux. It sounds like Linux has a little way to go before it is fully using the processing capabilities of Power architecture, and the sooner this happens the better for everyone involved.

_________________
Joshua Purcell


Top
   
 Post subject:
PostPosted: Tue Oct 03, 2006 7:40 am 
Offline

Joined: Thu Feb 16, 2006 8:10 pm
Posts: 98
Quote:
Quote:
Quote:
-Glibc PowerPC Optimizations, which will greatly increase the performance of memory-bound applications on Power architecture.
Remember http://www.freevec.org/ has optimizations for glibc (these will be ported to glibc proper soon) which far outweigh those on Gunnar's site. Not that he did a bad job, but he just cleaned up the crap; there's nothing like a bit of hand=vectorized code to speed things up though.
I not sure how these two projects fit in with each other. From what it sounds like, both are attempting to do the same thing (libfreevec is just further along). I did have a question about this project though:

-Does the end goal of the freevec project include getting these optimizations into glibc?

-Would it be beneficial to the freevec project to start a discussion over at Power.org about the work being done? I bring this up because there may be others there who would be willing to help in some way, and the only thing I've seen that is somewhat related to this is Gunnar's posts about his glib optimizations. There was an interesting discussion a month or two ago about this, and freevec was only mentioned in passing.

-I know there will always be multiple paths to the same goal (multiple Linux distributions for example), but what specifically are the difference between the work Gunnar has done and the freevec project? If they are the same thing working towards the same goal, then has there ever been any communication between the two projects?

I'm interested in this because all the talk I've heard about Power-optimized Linux. It sounds like Linux has a little way to go before it is fully using the processing capabilities of Power architecture, and the sooner this happens the better for everyone involved.
he's been away for a while so thats perhaps why it appears theres a way to go,but if you re-read the threads on this very site you will see that when their on a role it moves along very fast.

see
http://www.ppczone.org/forums/viewtopic ... sc&start=0 for the initial thread.

and the http://www.ppczone.org/forums/viewforum.php?f=23 section for more answers , the answer seems to be macOS used it, so linux will better it....

doing a search on this site for 'altivec' produces some good results and insights too

http://freevec.org/
"
Eventually, I hope to see the code being integrated in glibc so that the whole Linux system may benefit from AltiVec when run on the right hardware. The current version is released under the terms of the free LGPL license and I sincerly hope that it will enable many interested developers to give a performance boost to their applications for the PowerPC platform."


http://freevec.org/faq.php
"Q: There is already liboil, why don't you put your code there?
A: Actually I intend to, but not this particular code. The goal for liboil is slightly different, it offers its own API, and a whole lot of highly optimized routines to perform various algorithms. On the other hand, I wanted to optimize existing functions from GLIBC, libstring (which is heavily used in MySQL), etc. I do plan to write some code for liboil at a later stage, but not at this particular moment.
"


Top
   
Display posts from previous:  Sort by  
Post new topic  Reply to topic  [ 17 posts ] 

All times are UTC-06:00


Who is online

Users browsing this forum: No registered users and 17 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum

Search for:
Jump to:  
PowerDeveloper.org: Copyright © 2004-2012, Genesi USA, Inc. The Power Architecture and Power.org wordmarks and the Power and Power.org logos and related marks are trademarks and service marks licensed by Power.org.
All other names and trademarks used are property of their respective owners. Privacy Policy
Powered by phpBB® Forum Software © phpBB Group