I think Intel's attention is really somewhere else. I don't think Intel really cares so much about the Xeon or Pentium D or EM64T. These are such obvious half hearted attempts to try and keep up with AMD.
You would think as large and cash rich as Intel they should have little trouble producing chips that beat AMD. I think its obvious their concentration is really somewhere else.
The reality is that chip design for any company takes years. Intel had invested heavily in a single core, long pipeline, high clock rate strategy and was banking heavily on their process technology leadership to make it happen and dominate the market. Then everyone hit the wall at once, except that AMD wasn't playing that game because they knew they couldn't. So all of a sudden AMD was in a lead position and the big juggernaughts couldn't change course fast enough. PentiumD is a quick hack to "keep up with the Joneses", but a truly new chip design doesn't arrive until next year (even though it was begun 2-3 years ago).
Intel's EM64T is the same situation. They have been pouring money into an IA-64 future, and when AMD's x86-64 came out of left field and quickly became a hit, Intel was caught by surprise. The current EM64T solutions are quick hacks to answer the bullet point in AMD's feature list. EM64T isn't going away, instead it is waiting for a real implementation (coming in the new cores next year).
Intel still has the best process technology, and there is no sign of that changing. They aren't a company that is particularly leading edge in terms of new design directions and visionary advancements, quite probably for the very same reason. Once they get going in the right direction, however, look out. This is part of why Apple is moving to Intel (the other part being that Intel is hugely invested in the PC market, whereas IBM doesn't care about it).
Herm. Those 'twice as fast' figures for the Quad vs Brad's dual 2.5 figures bring a smile to my face. Very exciting to have a scene like that render in 17 secs!!! And with all that A.A on top!
Yeup. I know Max is PC only. I have it!
And I have dual Lightwave 8.x on the Mac/PC.
I also have the beautiful Vue 5 Infinite. And Poser 6...and...and...an'...
So. Polygon pushing power and rendering capability...'unlimited power' (Says in Emperor style voice...) are on my agenda.
A model with 250K polygons in it...with a few textures on top...I'd have thought a 7800 GT would have handled it with consumate ease.
I'd still like to see s Quadro vs 7800GT in action though...
The snooty Mac owner in me wouldn't mind a Quadro...but...at almost four times the cost? I question if it is four times more powerful...than a 7800GT.
25-50% better at CAD type stuff? For 300%/400% more?
I'd like to see a triangle vs triangle comparison....
Lemon Bon Bon
hi lemon, yeah sounds cool..!
hmm yeah i wonder what lightwave8 heavy users think about 7800gt vs quadro when doing 3d work. $1600 for the quadro
......17secs for global illumination and what was it, 8x AA? very nice. nice renderer, that luxology one.
edit: some interesting notes for lightwave 8.5
Please Note: LightWave version 8.5 has the following graphics card requirements:
(Required)
Full OpenGL Support (including OpenGL 2.0 support)
Latest drivers from chipset manufacturer
(Minimum)
nVidia FX 5200 series
ATI Radeon 9600
Minimum of 64 MB RAM (per display)
1024x768 minimum screen resolution
(Recommended)
nVIDIA Quadro FX 4000
256 MB RAM
1280x1024 screen resolution
Known issues:
-Some of the functions of the GLSL Hardware Shaders are currently having conflicts with ATI graphics cards. NewTek and ATI are working on this issue, so be sure to keep up with the latest drivers for your graphics card.
- OpenGL 2.0 is not fully supported in Mac OS X, so the option to use the GLSL Hardware Shaders is not available on the Mac at this time.
- OpenGL 2.0 is not fully supported in Mac OS X, so the option to use the GLSL Hardware Shaders is not available on the Mac at this time. [/i] [/B]
This is something that Apple could fix with an update. they don't have to wait 'till 10.5 Hopefully with the new machines and boards they will hurry this up.
that's why apple is "reluctant" to do certain things, they rarely make a half-assed attempt at it, though in some cases they initially only get 70% of the way and we're all like, wtf?
so yeah, nice powermac updates, but 7800gt, 6800ultra, a few quadro models, all available NOW, will be nice. plus GLSL support to round out full openGL 2.0 support now that you can get a super juicy Quadro with your dualie/quad powermacG5
here's off lightwave's page:
"GLSL Hardware Shader: View light fall offs, blended images on surfaces, procedurals, and even gradients in real-time in OpenGL viewports in Layout. This is all possible with OpenGL v2.0 Hardware Shaders on graphics cards which support the technology.
Geared toward giving you a more accurate preview of your final render, the GLSL Hardware Shaders can speed-up productivity for your projects."
i think you made a good choice. i know i was quite gung ho about 6600gt or 6800 7800 etc, but that's because i was desperate to try out half life 2, need for speed underground 2, and play some good ut2004 without going to an internet cafe. i am finding my 6600gt just nice for what i wanted, ie, playing some of the latest games on winxp. your 6600 with 256mb video memory with the dualcore 2.3ghz will be smooth and slick. all the best with your photoshopping
I just wanted the batter card for future proofing more then actual need. I'm a console gamer, and will be doing my gaming (slightly limited) on the PS3. I work on my machine, gaming on it would interfere with paying the bills
I should have my 2.3/23" LCD combo by Tuesday or Wednesday
Intel still has the best process technology, and there is no sign of that changing. They aren't a company that is particularly leading edge in terms of new design directions and visionary advancements, quite probably for the very same reason. Once they get going in the right direction, however, look out. This is part of why Apple is moving to Intel (the other part being that Intel is hugely invested in the PC market, whereas IBM doesn't care about it).
Yes it seems AMD has worked its way into becoming the nibble innovative company.
Will Intel be relegated to following AMD's path?
With IBM I don't quite understand what they are doing. They want to promote PPC. But are ignoring the largest propenent and maker of PPC machines.
Yes it seems AMD has worked its way into becoming the nibble innovative company.
Will Intel be relegated to following AMD's path?
With IBM I don't quite understand what they are doing. They want to promote PPC. But are ignoring the largest propenent and maker of PPC machines.
What people are ignoring is that Intel made a remarkable turn around, especially for such a large company. They literally abondoned their entire R&D that had cost them billions over years, their entire furure chip design direction, and made a 90 degree turn in everything. They were able to do this in less than 1 year.
Anyone who thinks that Intel won't regain the lead is not thinking clearly.
As for IBM, who knows? It's difficult to figure out what they are intending to do. The have their own schedule to follow, and Apple's needs just had to wait for it to get where IBM wanted it to be at any given time.
What people are ignoring is that Intel made a remarkable turn around, especially for such a large company. They literally abondoned their entire R&D that had cost them billions over years, their entire furure chip design direction, and made a 90 degree turn in everything. They were able to do this in less than 1 year.
Anyone who thinks that Intel won't regain the lead is not thinking clearly.
Well said. They are currently just building momentum, and when they arrive... look out. I would say that it wasn't just a 1 year course change though, it began internally longer ago than that and they are still trying to figure out where exactly they are going.
Quote:
As for IBM, who knows? It's difficult to figure out what they are intending to do. The have their own schedule to follow, and Apple's needs just had to wait for it to get where IBM wanted it to be at any given time.
Well its pretty simple, really. Apple is selling 1-5 million machines per year and that isn't likely to change significantly in the foreseeable future. And to do this they need at least 2 different chip designs (laptop vs. desktop), one of which IBM has no other customers for at the moment. On the other hand IBM has Microsoft who wants to sell 10+ million game systems a year, Sony who wants to cell (sic) 10+ million game systems per year plus various other gadgets on essentially the same tech, plus Toshiba who wants to sell all sorts of TVs and other gadgets in huge volumes... and those guys will take care of most of their own fabbing and production (the expensive and low margin end of things). IBM can continue to crank out new designs and focus on higher margin lower volume production. Plus they don't have to listen to Steve ranting about how they aren't delivering.
I know that everybody here is all about the graphics and such. Being a Logic Pro user, I feel a little left out in this thread. My question is what is this quad G5 going to do for people using Logic. Is Logic going to take advantage of all of this power or is it a waste of money. I want to get a bare-bones quad and upgrade the memory myself, but if there is no gains to be seen in Logic, what is the point? Thanks.
I know that everybody here is all about the graphics and such. Being a Logic Pro user, I feel a little left out in this thread. My question is what is this quad G5 going to do for people using Logic. Is Logic going to take advantage of all of this power or is it a waste of money. I want to get a bare-bones quad and upgrade the memory myself, but if there is no gains to be seen in Logic, what is the point? Thanks.
It depends on what you do. If all you need is two channels and straightfoward editing, then, no, the Quad is not for you.
I do this for a living. I regularly push 24+ tracks with a ton of automation and of course some DSP on most tracks (if needed). I don't plan on upgrading past the last PowerMac compatible Logic version for at least 4 years (from today), so I need something that will take a beating for that long. Plus, as with any studio, you need to have the capibilties to do anything. You must be prepared for the eventuality that a client will want this, that and the other thing at any given time. So, I'd like a machine that I can count on to do that for me for a while. If the quad can offer me more than the 2.7 GHz G5 when it comes to DSP and softsynths, then it is my machine.
I do this for a living. I regularly push 24+ tracks with a ton of automation and of course some DSP on most tracks (if needed). I don't plan on upgrading past the last PowerMac compatible Logic version for at least 4 years (from today), so I need something that will take a beating for that long. Plus, as with any studio, you need to have the capibilties to do anything. You must be prepared for the eventuality that a client will want this, that and the other thing at any given time. So, I'd like a machine that I can count on to do that for me for a while. If the quad can offer me more than the 2.7 GHz G5 when it comes to DSP and softsynths, then it is my machine.
I use Logic Pro as well as Pro Tools, so if that's what you do, I say go for it.
I'm getting one in January as I like to give a bit of time when a completely new machine comes out.
With the Quad, you can most likely push 48 tracks. But you would want at least 2GB RAM. I plan to get 4.
Yep, I already checked Crucial for pricing on RAM. I'll be getting at least 4 gigs myself. I run a PT HD3 rig at the other studio I work at, but I prefer Logic for my studio. The $30,000 price tag on that PT rig just isn't cost effective for me, plus I like using the converters and pres I currently have.
Yep, I already checked Crucial for pricing on RAM. I'll be getting at least 4 gigs myself. I run a PT HD3 rig at the other studio I work at, but I prefer Logic for my studio. The $30,000 price tag on that PT rig just isn't cost effective for me, plus I like using the converters and pres I currently have.
They're both good for what they do. But they don't do exactly the same things.
With IBM I don't quite understand what they are doing. They want to promote PPC. But are ignoring the largest propenent and maker of PPC machines.
While they aren't necessarily proponents, but Microsoft, Sony and Nintendo will likely produce per-company as many PPC machines as Apple did, though probably not at as high of a margin.
Anyone who thinks that Intel won't regain the lead is not thinking clearly.
AMD executives seem to definitively state Intel won't regain its lead.
While of course I know about IBM's race to win the consumer gaming platform and the markets profitability.
What makes this confusing is IBM's propaganda push for "Power Everywhere" campaign.
Quote:
Industry analyst Jonathan Eunice of Illuminata in a perspective entitled ?Power Not so Everywhere? said, ??if the combination of IBM, Freescale, etc. could not meet the rather mainstream needs of a long-term, premier partner that buys millions of units every year, that does not speak well for the combination of the architecture [Power architecture] and its ecosystem.
The AMD are doing fine in benchmarks but unfortunately the majority of clients are still ordering Intel based servers likely 5 to 1(when there are AMD options).
AMD is going to need more than "Hey we're faster" to dent Intel's armor.
Comments
Originally posted by TenoBell
I think Intel's attention is really somewhere else. I don't think Intel really cares so much about the Xeon or Pentium D or EM64T. These are such obvious half hearted attempts to try and keep up with AMD.
You would think as large and cash rich as Intel they should have little trouble producing chips that beat AMD. I think its obvious their concentration is really somewhere else.
The reality is that chip design for any company takes years. Intel had invested heavily in a single core, long pipeline, high clock rate strategy and was banking heavily on their process technology leadership to make it happen and dominate the market. Then everyone hit the wall at once, except that AMD wasn't playing that game because they knew they couldn't. So all of a sudden AMD was in a lead position and the big juggernaughts couldn't change course fast enough. PentiumD is a quick hack to "keep up with the Joneses", but a truly new chip design doesn't arrive until next year (even though it was begun 2-3 years ago).
Intel's EM64T is the same situation. They have been pouring money into an IA-64 future, and when AMD's x86-64 came out of left field and quickly became a hit, Intel was caught by surprise. The current EM64T solutions are quick hacks to answer the bullet point in AMD's feature list. EM64T isn't going away, instead it is waiting for a real implementation (coming in the new cores next year).
Intel still has the best process technology, and there is no sign of that changing. They aren't a company that is particularly leading edge in terms of new design directions and visionary advancements, quite probably for the very same reason. Once they get going in the right direction, however, look out. This is part of why Apple is moving to Intel (the other part being that Intel is hugely invested in the PC market, whereas IBM doesn't care about it).
Originally posted by Lemon Bon Bon
Herm. Those 'twice as fast' figures for the Quad vs Brad's dual 2.5 figures bring a smile to my face. Very exciting to have a scene like that render in 17 secs!!! And with all that A.A on top!
Yeup. I know Max is PC only. I have it!
And I have dual Lightwave 8.x on the Mac/PC.
I also have the beautiful Vue 5 Infinite.
So. Polygon pushing power and rendering capability...'unlimited power' (Says in Emperor style voice...) are on my agenda.
A model with 250K polygons in it...with a few textures on top...I'd have thought a 7800 GT would have handled it with consumate ease.
I'd still like to see s Quadro vs 7800GT in action though...
The snooty Mac owner in me wouldn't mind a Quadro...but...at almost four times the cost? I question if it is four times more powerful...than a 7800GT.
25-50% better at CAD type stuff? For 300%/400% more?
I'd like to see a triangle vs triangle comparison....
Lemon Bon Bon
hi lemon, yeah sounds cool..!
hmm yeah i wonder what lightwave8 heavy users think about 7800gt vs quadro when doing 3d work. $1600 for the quadro
......17secs for global illumination and what was it, 8x AA? very nice. nice renderer, that luxology one.
edit: some interesting notes for lightwave 8.5
Please Note: LightWave version 8.5 has the following graphics card requirements:
(Required)
Full OpenGL Support (including OpenGL 2.0 support)
Latest drivers from chipset manufacturer
(Minimum)
nVidia FX 5200 series
ATI Radeon 9600
Minimum of 64 MB RAM (per display)
1024x768 minimum screen resolution
(Recommended)
nVIDIA Quadro FX 4000
256 MB RAM
1280x1024 screen resolution
Known issues:
-Some of the functions of the GLSL Hardware Shaders are currently having conflicts with ATI graphics cards. NewTek and ATI are working on this issue, so be sure to keep up with the latest drivers for your graphics card.
- OpenGL 2.0 is not fully supported in Mac OS X, so the option to use the GLSL Hardware Shaders is not available on the Mac at this time.
Originally posted by sunilraman
- OpenGL 2.0 is not fully supported in Mac OS X, so the option to use the GLSL Hardware Shaders is not available on the Mac at this time. [/i] [/B]
This is something that Apple could fix with an update. they don't have to wait 'till 10.5 Hopefully with the new machines and boards they will hurry this up.
so yeah, nice powermac updates, but 7800gt, 6800ultra, a few quadro models, all available NOW, will be nice. plus GLSL support to round out full openGL 2.0 support now that you can get a super juicy Quadro with your dualie/quad powermacG5
here's off lightwave's page:
"GLSL Hardware Shader: View light fall offs, blended images on surfaces, procedurals, and even gradients in real-time in OpenGL viewports in Layout. This is all possible with OpenGL v2.0 Hardware Shaders on graphics cards which support the technology.
Geared toward giving you a more accurate preview of your final render, the GLSL Hardware Shaders can speed-up productivity for your projects."
GLSL off:
GLSL on:
Originally posted by sunilraman
i think you made a good choice. i know i was quite gung ho about 6600gt or 6800 7800 etc, but that's because i was desperate to try out half life 2, need for speed underground 2, and play some good ut2004 without going to an internet cafe. i am finding my 6600gt just nice for what i wanted, ie, playing some of the latest games on winxp. your 6600 with 256mb video memory with the dualcore 2.3ghz will be smooth and slick. all the best with your photoshopping
I just wanted the batter card for future proofing more then actual need. I'm a console gamer, and will be doing my gaming (slightly limited) on the PS3. I work on my machine, gaming on it would interfere with paying the bills
I should have my 2.3/23" LCD combo by Tuesday or Wednesday
Intel still has the best process technology, and there is no sign of that changing. They aren't a company that is particularly leading edge in terms of new design directions and visionary advancements, quite probably for the very same reason. Once they get going in the right direction, however, look out. This is part of why Apple is moving to Intel (the other part being that Intel is hugely invested in the PC market, whereas IBM doesn't care about it).
Yes it seems AMD has worked its way into becoming the nibble innovative company.
Will Intel be relegated to following AMD's path?
With IBM I don't quite understand what they are doing. They want to promote PPC. But are ignoring the largest propenent and maker of PPC machines.
Originally posted by TenoBell
Yes it seems AMD has worked its way into becoming the nibble innovative company.
Will Intel be relegated to following AMD's path?
With IBM I don't quite understand what they are doing. They want to promote PPC. But are ignoring the largest propenent and maker of PPC machines.
What people are ignoring is that Intel made a remarkable turn around, especially for such a large company. They literally abondoned their entire R&D that had cost them billions over years, their entire furure chip design direction, and made a 90 degree turn in everything. They were able to do this in less than 1 year.
Anyone who thinks that Intel won't regain the lead is not thinking clearly.
As for IBM, who knows? It's difficult to figure out what they are intending to do. The have their own schedule to follow, and Apple's needs just had to wait for it to get where IBM wanted it to be at any given time.
Originally posted by melgross
What people are ignoring is that Intel made a remarkable turn around, especially for such a large company. They literally abondoned their entire R&D that had cost them billions over years, their entire furure chip design direction, and made a 90 degree turn in everything. They were able to do this in less than 1 year.
Anyone who thinks that Intel won't regain the lead is not thinking clearly.
Well said. They are currently just building momentum, and when they arrive... look out. I would say that it wasn't just a 1 year course change though, it began internally longer ago than that and they are still trying to figure out where exactly they are going.
As for IBM, who knows? It's difficult to figure out what they are intending to do. The have their own schedule to follow, and Apple's needs just had to wait for it to get where IBM wanted it to be at any given time.
Well its pretty simple, really. Apple is selling 1-5 million machines per year and that isn't likely to change significantly in the foreseeable future. And to do this they need at least 2 different chip designs (laptop vs. desktop), one of which IBM has no other customers for at the moment. On the other hand IBM has Microsoft who wants to sell 10+ million game systems a year, Sony who wants to cell (sic) 10+ million game systems per year plus various other gadgets on essentially the same tech, plus Toshiba who wants to sell all sorts of TVs and other gadgets in huge volumes... and those guys will take care of most of their own fabbing and production (the expensive and low margin end of things). IBM can continue to crank out new designs and focus on higher margin lower volume production. Plus they don't have to listen to Steve ranting about how they aren't delivering.
Originally posted by jaffi
I know that everybody here is all about the graphics and such. Being a Logic Pro user, I feel a little left out in this thread. My question is what is this quad G5 going to do for people using Logic. Is Logic going to take advantage of all of this power or is it a waste of money. I want to get a bare-bones quad and upgrade the memory myself, but if there is no gains to be seen in Logic, what is the point? Thanks.
It depends on what you do. If all you need is two channels and straightfoward editing, then, no, the Quad is not for you.
Originally posted by jaffi
I do this for a living. I regularly push 24+ tracks with a ton of automation and of course some DSP on most tracks (if needed). I don't plan on upgrading past the last PowerMac compatible Logic version for at least 4 years (from today), so I need something that will take a beating for that long. Plus, as with any studio, you need to have the capibilties to do anything. You must be prepared for the eventuality that a client will want this, that and the other thing at any given time. So, I'd like a machine that I can count on to do that for me for a while. If the quad can offer me more than the 2.7 GHz G5 when it comes to DSP and softsynths, then it is my machine.
I use Logic Pro as well as Pro Tools, so if that's what you do, I say go for it.
I'm getting one in January as I like to give a bit of time when a completely new machine comes out.
With the Quad, you can most likely push 48 tracks. But you would want at least 2GB RAM. I plan to get 4.
Originally posted by jaffi
Yep, I already checked Crucial for pricing on RAM. I'll be getting at least 4 gigs myself. I run a PT HD3 rig at the other studio I work at, but I prefer Logic for my studio. The $30,000 price tag on that PT rig just isn't cost effective for me, plus I like using the converters and pres I currently have.
They're both good for what they do. But they don't do exactly the same things.
- OpenGL 2.0 is not fully supported in Mac OS X, so the option to use the GLSL Hardware Shaders is not available on the Mac at this time.
PULL your finger OUT Apple...!
Lemon Bon Bon
Originally posted by TenoBell
With IBM I don't quite understand what they are doing. They want to promote PPC. But are ignoring the largest propenent and maker of PPC machines.
While they aren't necessarily proponents, but Microsoft, Sony and Nintendo will likely produce per-company as many PPC machines as Apple did, though probably not at as high of a margin.
Anyone who thinks that Intel won't regain the lead is not thinking clearly.
AMD executives seem to definitively state Intel won't regain its lead.
While of course I know about IBM's race to win the consumer gaming platform and the markets profitability.
What makes this confusing is IBM's propaganda push for "Power Everywhere" campaign.
Industry analyst Jonathan Eunice of Illuminata in a perspective entitled ?Power Not so Everywhere? said, ??if the combination of IBM, Freescale, etc. could not meet the rather mainstream needs of a long-term, premier partner that buys millions of units every year, that does not speak well for the combination of the architecture [Power architecture] and its ecosystem.
Originally posted by TenoBell
AMD executives seem to definitively state Intel won't regain its lead.
That sort of thing reminds me of fake wrestling.
I imagine they will try as hard as they can, but I don't see how they can say for certain.
Originally posted by TenoBell
AMD executives seem to definitively state Intel won't regain its lead.
Umm, that's their job. They aren't going to open with we're incompetent and going to let Intel have it back.
AMD is going to need more than "Hey we're faster" to dent Intel's armor.