If a card can truly saturate an x16, it would be severely throttled at x8. A 7800 *might* be called as saturating an x2 lane system. I doubt that an 8600 or 8800 can consume four times that data rate.
Again. Were did I ever say that a 16X lane could be fully saturated? WTF?
It's getting harder to find reliable SLI lane tests on the internet, and this poster actually cleared up a lot for me as to why I can not find anything that looks reliable anymore.
Quote:
I think thats mistaken. Since there are 32x boards out... plus when you factor in all the lanes for the other onboard devices that use the PCI-e bus... you would still be over 20 lanes with a single 16x and 4 slot on the mainboard...
Read the PCIe spec. You are confusing slot lenght (1x, 2x, 4x, 16x) with the total available lanes which is not surprising since most review sites do it too. While 8X lenght slots are valid you will not see them used on any consumer MB's.
While how the lanes are distributed amoung the slots varies some from maker to maker a typical triple spaced, dual 16X slot PCIe MB will have a 1X and a 2X slot in between the 16X slots. In single card config the lanes will be distributed as 16 to PCIe1 which is normally the slot closest to the CPU, no lanes to the 1X (in other words disabled), 2 lanes to the 2X slot and 2 lanes to PCIe4 which is the 2nd 16X slot.
When configured for SLI or CF that same MD will have 8 lanes going to each 16X PCIe slot (PCIe1 and PCIe4), 1 lane to the 1X slot, 2 lanes to the 2X slot
and 1 lane unused.
There are of course other configurations of slot sizes used with different distributions of the available lanes to them but the above is fairly common.
Viper
I think you can see how much confusion could potentially be made from that.
And for those who do like to game NVidia just broke new ground, and in turn puts Apple two generations behind in graphics capabilities.
Using three GPU's the 3-way NVIDIA SLI takes extreme gaming to a whole new level.
Friday, 14 December 2007
Extreme gaming just got a whole lot better. NVIDIA has released and extended its SLI technology, which enables the use of multiple graphics processing units (GPUs) on a single computer, allowing up to three GeForce graphics cards to be used in a single machine.
Now the graphics-intensive titles, such as Call of Duty 4, Company of Heroes Opposing Fronts, Enemy Territory: Quake Wars, and Unreal Tournament 3, can be played at the highest resolution possible, with all the graphics settings cranked to the max, and antialiasing applied for the first time. Also, for the serious user in production, the rendering speed generated out of the use of the SLI is nothing less than brilliant.
NVIDIA?s new 3-way SLI delivers up to a 2.8x performance increase over a single GPU system, giving high-end gamers 60 frames per second at resolutions as high as 2560x1600 and with 8x antialiasing. 3-way SLI technology means there'll be no more dialing back the image quality settings on the newest PC games. For example, gamers with 3-way SLI can play 'Call of Duty 4' at high resolutions such as 1920x1200 with all the advanced DirectX 10 effects such as motion blur, ambient occlusion, and soft shadows turned on. Its the full experience that the producers and the artists wanted the game players to see.
The heart of a 3-way SLI system is an NVIDIA nForce® 680 SLI MCP motherboard and three GeForce 8800 GTX or GeForce 8800 Ultra graphics cards. With 3-way SLI, gamers can harness the power of 384 stream processors, a 110+ gigatexel per second texture fill rate, and over two gigabytes of graphics memory for no-compromise gaming performance.
3-way SLI gives gamers the flexibility to scale their graphics processing power with one, two, or three GeForce GPUs, depending on their desired price and system configuration. 3-way SLI systems are available from leading gaming PC system builders and the components needed to build your own 3-way SLI system are available from leading retailers.
--------------------
And for the production Artist it looks like real time shader, texture, bump mapping, and light, rendering on the fly of a 11 million polly model is getting closer at hand.
Minimum Power Supply is 1100W. Somehow I'd imagine the design of the motherboard needing 3 full PCIe x16 slots makes this not materially possible, even if the drivers were available.
I am beginning to think new Mac Pros will be announced after Macworld, but sooner than NAB or WWDC. I assume they will be in new enclosures at that time.
Steve will probably spend half of the keynote on Leopard (how many they've sold, update today, etc.), new Mac stores, revenue, another look at iLife, iTunes with a new update. Then if they introduce some new product, a consumer/prosumer device, that will eat up the rest of the time. Then a close with an up and coming musician. Unfortunately, we won't be seeing new Pro macs at MacWorld.
I think you are mistaken. Expect the 8-core Mac Pro to have a staring role in the January 15th SteveNote.
Quote:
Originally Posted by emig647
If you're saying apple is going to wait until NAB or "photo shows" I highly disagree. The next event after MWSF is usually in March or April. Apple can't go that long without releasing. Apple has updated these machines on normal days in the passed, but I don't see any reason for apple to release these machines between now and MWSF... and I can't imagine they go much passed MWSF.
I will be in shock if it's not January 15.
Quote:
Originally Posted by emig647
After macworld would be insanity... The mac pro specs haven't been touched since August 2006... Why wouldn't they announce it at MacWorld if it's ready? It's free Press.
Zactly.
Quote:
Originally Posted by emig647
Don't gotta apologize. It's all speculation. Apple could announce any day, any time they want. They have done it many many times in the past.
You're right, it is insanity they have gone this long. I personally don't get it. They HAVE the cpus. What else is there to wait for? Graphics cards? Motherboards? If dell has the motherboards I'm sure apple does too. All I can think of is graphics.
They had to wait for the Intel Penryn Developer Kit which was only released week before last so they could tweak Leopard for Penryn efficiency adding all the new multimedia hooks to 10.5.2 so the system software is in tune with the new Penryn, Stoakley-Seaburg hardware. They also had to wait for the supply of 3.2 GHz Harpertown Quad Core Penryns - the last to be delivered in significant quantity - to ramp up. I don't see any reason to think this next gen 2008 Mac Pro is anything but on time as soon as it could have possibly been delivered. And I also don't see how it's not going to be one of the centerpieces of Steves's message January 15th. I.E. All the pieces to the new Mac Pro will be in place by January 15, 2008. There's absolutely no reason to think it's going to be later than that.
Its going to purely be a function of how well they are yielding because, as you point out, it doesn't look like it'll be driven by competition from AMD. The mid-3 GHz range seems achievable though.
*Nods.
Quote:
Nehalem sounds like the more exciting chip because of the memory controller stuff.
Well, there's been hype about this in AMD chips for years...
The fact that its a whole new microarchitecture is the exciting part. Everybody has been clamoring for on-chip memory controllers for ages, but really I'm not sure how big a win it is in terms of performance in desktop systems. It should be a significant cost and TDP win since there are fewer parts, and it should reduce the amount of L2 cache they are obliged to include... but in terms of real-world performance due to the on-die memory controller alone, I don't think it'll live up to the hype around the issue. The new chip interconnect will be a nice addition at the MacPro level -- finally a competitor to HyperTransport... but again, on most software is it going to be a big win? The return of HyperThreading is also interesting for software that needs lots of threads, but really... with 4+ cores already are most users going to see a difference? Perhaps as software evolves over the next few years, but on the day these things ship I doubt it.
Well, I was wondering why I should wait for Nehalem when a quad or 8 way Penryn would surely do just fine. And surely, in light of current Mac Pros... Grr...the GPU is the more exciting prospect? A 9900 will seem very special after an Ati 1900 I would have thought? And in terms of 3D...Lightwave...the odd game, lots of Photoshopping, a quad or 8 core with a decent GPU would do me just fine.
Even most consumer Apple apps only seem to use 1 and a half cores half the time, heh. I think the big win is going to have to come from Software makers over the next few years.
Didn't programmers use to have to consider the chips in the Amiga? Are programmers just aiming for the 'middle ground' with their code? Are they writing lazy 'mono-thread' code..? Or do they just need to wait until Quad/Duo core reaches critical mass before they change the 'mhz' culture code?
I guess you can provide insight into that...
The GPU core tightly coupled to the CPU may be the most interesting technology. Intel has come a long way with their integrated GPUs, and I would guess that this thing is the next generation after the current X3100 and it will benefit from high speed interconnects with the CPU and the CPU's on-chip memory controller. There is no doubt that all the high-end GPU fans will slag this mercilessly just like they do the X3100 and its predecessors... but the fact remains that for the iMac and MacBook this will be a big step up. Possibly a very big step up.
Well, it seems Intel are coming round the mountain with GPUs. If they're going to exceed the x3100 then surely you'll have GPU performance in excess of a Nintendo Wii? Hey, don't knock it folks. Nintendo have bowed out the arms race and seem to be doing ok... But that's one thing... Apple's Mac Pro's should not be bowing out the gpu arms race...heh. Jan 15th seems ages away...
Quote:
What's Nehalem going to offer worth waiting a year for?
Yeah, that's my point. With Penyrn the transition to 45nm is a win, along with various architectural refinements... Nehalem is going to bring a bunch of new forward-looking stuff, but its not clear how big an improvement will be delivered on day 1.
But like many forward thinking things...we're not going to see the benefit until Quad core and maybe even 8core are mainstream? Most software seems to be only just emerging from the dark ages in terms of threading?
Looks at Photoshop. Apple demo'd funhouse as a real time filters app. Yet, did Adobe put that realtime functionality in? Nooooope.
Software makers have to get on board. Maybe it will be the 'Coco' Mac developing 'young guns' that will show what the Mac can really do now we have Leopard out.
Addendum: I just came across an Intel page discussing Nehalem, and the stuff it talks about is mostly of interest to Intel and PC makers. It is aimed at created a chip family that scales from the low end to the high end, from the low-power to the high-power. There is no discussion of specific performance enhancements or new technologies that should make a buyer wait for it. Laptops and AIO machines will benefit the most, it looks like.
A further reason for Apple's laptop sales curve to increase.
But a dinosaur like me wants a honking quad/8 core gpu monster. Hmmm.
A commendable post, Programmer. Thankyou for doing the leg work. You've certainly given me leave to buy the Mac Pro without worrying about being left behind by Nehalem. After an 8 year wait...I think Feb' might be the time to 'come home'. *Looks to the horizon.
PS. I've kinda chatted inbetween your quote. Kinda hard to see in blue...
Thanks for pointing it out, I wouldn't have noticed otherwise. I think you need to learn to use the
Quote:
UBB quoting directives
.
The Amiga had multiple special purpose chips, but only one "main processor". The special purpose chips were fed sequences of commands, but these weren't the same as a fully general purpose instruction set like is found in a full fledged CPU. The first mainstream desktop machines with multiple processors started showing up in the mid-90s, although there had been many specialized machines before that (including add-on boards, super computers, etc.). Some of the early attempts provided an extra processor specifically to do graphics (the first attempts at a graphics coprocessor).
Multi-threaded programming is hard, and the tools & languages currently provided for developers to do their thing aren't very good. Frankly, you probably don't want a lot of the programmers out their writing multi-threaded programs directly because it is easy to create poorly performing, unstable multi-threaded software. This is why sophisticated system software services like OpenGL, QuickTime, CoreAnimation, CoreAudio, etc. are extremely valuable... Apple takes care of the hard parts and let's the application developers get their apps done faster and simpler. Over time the tools will improve and make the task of creating multi-threaded software easier and safer. In the meantime you have to wait for the more highly skilled developers to "do it the hard way". Only relatively recently have multi-core chips become common enough to really be worthy of the attention of most developers... previously it was only a few companies who targeted these machines.
I think you are mistaken. Expect the 8-core Mac Pro to have a staring role in the January 15th SteveNote.
And I also don't see how it's not going to be one of the centerpieces of Steves's message January 15th. I.E. All the pieces to the new Mac Pro will be in place by January 15, 2008. There's absolutely no reason to think it's going to be later than that.
The Amiga had multiple special purpose chips, but only one "main processor". The special purpose chips were fed sequences of commands, but these weren't the same as a fully general purpose instruction set like is found in a full fledged CPU. The first mainstream desktop machines with multiple processors started showing up in the mid-90s, although there had been many specialized machines before that (including add-on boards, super computers, etc.). Some of the early attempts provided an extra processor specifically to do graphics (the first attempts at a graphics coprocessor).
Multi-threaded programming is hard, and the tools & languages currently provided for developers to do their thing aren't very good. Frankly, you probably don't want a lot of the programmers out their writing multi-threaded programs directly because it is easy to create poorly performing, unstable multi-threaded software. This is why sophisticated system software services like OpenGL, QuickTime, CoreAnimation, CoreAudio, etc. are extremely valuable... Apple takes care of the hard parts and let's the application developers get their apps done faster and simpler. Over time the tools will improve and make the task of creating multi-threaded software easier and safer. In the meantime you have to wait for the more highly skilled developers to "do it the hard way". Only relatively recently have multi-core chips become common enough to really be worthy of the attention of most developers... previously it was only a few companies who targeted these machines.
Noted.
I guess we're just going to have to be patient with the multi-threading thing. I would have thought with programmers working on the multi-chip 360 and the PS3 consoles that design of the chips/architecture would create a mainstream environment for threading to go mainstream...especially as they move to squeeze every drop of performance out of them. I would have thought that will help or contribute to the threading gene pool in the next 3 years as these beasts hit critical mass in their millions upon millions. The incentive to multi-thread, financially will be there? And these developers can feed the techniques back into the land of computers?
Until then, at least I can be thankful that Lightwave will take advantage of as many cores as one can throw at it (according to one Newtek guy I emailed...) Plus, I guess I can play the odd game of Eve Online, some Photoshop work...all while LWave renders in the background...and do all this using Leopard's Spaces. I guess it's not all about individual apps. Sometimes the user can just multitask...that's one way to get the most out of yeh cores in the meantime... Heh.
I think you are mistaken. Expect the 8-core Mac Pro to have a staring role in the January 15th SteveNote. I will be in shock if it's not January 15.
Prepare to be in shock. See here. Although, if true, that certainly cannot be laid at Apple's door.
I still would be extremely disappointed if there is not a least a refresh of the dual core version of Mac Pro during MWSF. The only reason for not refreshing it would be if it is slowly being phased out of their product line.
Prepare to be in shock. See here. Although, if true, that certainly cannot be laid at Apple's door.
I still would be extremely disappointed if there is not a least a refresh of the dual core version of Mac Pro during MWSF. The only reason for not refreshing it would be if it is slowly being phased out of their product line.
Those quads delayed are not Xeons, they are desktop quads and Apple doesn't use any desktop cpus from Intel.
Xeon penryns have been available since late november, most manufacturers have announced workstations and servers using those new penryn Xeons, but most will only be available late december or january.
Apple may be waiting to have enough new chips (and i5400 chipsets), and sell much of their current inventory before releasing a new Mac Pro.
Especially if the line up comes with 8 cores all the way, I really think that Apple will announce it at MWSF (8 cores starting at $2200! since those new chips are priced like the old dual-core woodcrest ones).
The question is: will they release something in between the dual-core iMac and a 8-core Mac Pro in the desktop segment?
Xeon penryns have been available since late november, most manufacturers have announced workstations and servers using those new penryn Xeons, but most will only be available late december or january.
Those are differing ideas of available. I'm sure the chip has been out in sample quantities but I haven't heard of anyone shipping any machines at all.
Quote:
Apple may be waiting to have enough new chips (and i5400 chipsets), and sell much of their current inventory before releasing a new Mac Pro.
Especially if the line up comes with 8 cores all the way, I really think that Apple will announce it at MWSF (8 cores starting at $2200! since those new chips are priced like the old dual-core woodcrest ones).
The question is: will they release something in between the dual-core iMac and a 8-core Mac Pro in the desktop segment?
Personally, I hope so. Apple used to offer PowerMacs as low as $1499.
Prepare to be in shock. See here. Although, if true, that certainly cannot be laid at Apple's door.
I still would be extremely disappointed if there is not a least a refresh of the dual core version of Mac Pro during MWSF. The only reason for not refreshing it would be if it is slowly being phased out of their product line.
This is why AMD can't die out. If they do, Intel have no reason to make an effort. They only need to be better than their competition to make the sale.
Prepare to be in shock. See here. Although, if true, that certainly cannot be laid at Apple's door.
I still would be extremely disappointed if there is not a least a refresh of the dual core version of Mac Pro during MWSF. The only reason for not refreshing it would be if it is slowly being phased out of their product line.
And what do you think a refresh of dual core would look like?\
Comments
If a card can truly saturate an x16, it would be severely throttled at x8. A 7800 *might* be called as saturating an x2 lane system. I doubt that an 8600 or 8800 can consume four times that data rate.
Again. Were did I ever say that a 16X lane could be fully saturated? WTF?
Also your X2 is an X2 lane, not a Dual GPU on one card like the X2 I am talking about. Just to clarify.
X2 and x2 mean different things. The caps do mean very specific things, and I do understand that.
Again. Were did I ever say that a 16X lane could be fully saturated? WTF?
"but you can definitely saturate a X16 lane with any New Nvidia Card from the 8600 series and up, "
fully saturate was redundant on my part.
"but you can definitely saturate a X16 lane with any New Nvidia Card from the 8600 series and up, "
A misuse of the terminology, (I just wrote the wrong speed down I meant 8x isn't enough) and I have since restated my meaning at least once.
I think thats mistaken. Since there are 32x boards out... plus when you factor in all the lanes for the other onboard devices that use the PCI-e bus... you would still be over 20 lanes with a single 16x and 4 slot on the mainboard...
Read the PCIe spec. You are confusing slot lenght (1x, 2x, 4x, 16x) with the total available lanes which is not surprising since most review sites do it too. While 8X lenght slots are valid you will not see them used on any consumer MB's.
While how the lanes are distributed amoung the slots varies some from maker to maker a typical triple spaced, dual 16X slot PCIe MB will have a 1X and a 2X slot in between the 16X slots. In single card config the lanes will be distributed as 16 to PCIe1 which is normally the slot closest to the CPU, no lanes to the 1X (in other words disabled), 2 lanes to the 2X slot and 2 lanes to PCIe4 which is the 2nd 16X slot.
When configured for SLI or CF that same MD will have 8 lanes going to each 16X PCIe slot (PCIe1 and PCIe4), 1 lane to the 1X slot, 2 lanes to the 2X slot
and 1 lane unused.
There are of course other configurations of slot sizes used with different distributions of the available lanes to them but the above is fairly common.
Viper
I think you can see how much confusion could potentially be made from that.
and boards with on board pci-x slots use x4 lanes for the pci-e to pci-x chip.
x4 and up raid cards are also becomeing a big things as well.
and boards with on board pci-x slots use x4 lanes for the pci-e to pci-x chip.
PC's never stopped having PCI-X when PCI-E came out. Just Mac's.
And for those who do like to game NVidia just broke new ground, and in turn puts Apple two generations behind in graphics capabilities.
Using three GPU's the 3-way NVIDIA SLI takes extreme gaming to a whole new level.
Friday, 14 December 2007
Extreme gaming just got a whole lot better. NVIDIA has released and extended its SLI technology, which enables the use of multiple graphics processing units (GPUs) on a single computer, allowing up to three GeForce graphics cards to be used in a single machine.
Now the graphics-intensive titles, such as Call of Duty 4, Company of Heroes Opposing Fronts, Enemy Territory: Quake Wars, and Unreal Tournament 3, can be played at the highest resolution possible, with all the graphics settings cranked to the max, and antialiasing applied for the first time. Also, for the serious user in production, the rendering speed generated out of the use of the SLI is nothing less than brilliant.
NVIDIA?s new 3-way SLI delivers up to a 2.8x performance increase over a single GPU system, giving high-end gamers 60 frames per second at resolutions as high as 2560x1600 and with 8x antialiasing. 3-way SLI technology means there'll be no more dialing back the image quality settings on the newest PC games. For example, gamers with 3-way SLI can play 'Call of Duty 4' at high resolutions such as 1920x1200 with all the advanced DirectX 10 effects such as motion blur, ambient occlusion, and soft shadows turned on. Its the full experience that the producers and the artists wanted the game players to see.
The heart of a 3-way SLI system is an NVIDIA nForce® 680 SLI MCP motherboard and three GeForce 8800 GTX or GeForce 8800 Ultra graphics cards. With 3-way SLI, gamers can harness the power of 384 stream processors, a 110+ gigatexel per second texture fill rate, and over two gigabytes of graphics memory for no-compromise gaming performance.
3-way SLI gives gamers the flexibility to scale their graphics processing power with one, two, or three GeForce GPUs, depending on their desired price and system configuration. 3-way SLI systems are available from leading gaming PC system builders and the components needed to build your own 3-way SLI system are available from leading retailers.
--------------------
And for the production Artist it looks like real time shader, texture, bump mapping, and light, rendering on the fly of a 11 million polly model is getting closer at hand.
Minimum Power Supply is 1100W. Somehow I'd imagine the design of the motherboard needing 3 full PCIe x16 slots makes this not materially possible, even if the drivers were available.
I am beginning to think new Mac Pros will be announced after Macworld, but sooner than NAB or WWDC. I assume they will be in new enclosures at that time.
Steve will probably spend half of the keynote on Leopard (how many they've sold, update today, etc.), new Mac stores, revenue, another look at iLife, iTunes with a new update. Then if they introduce some new product, a consumer/prosumer device, that will eat up the rest of the time. Then a close with an up and coming musician. Unfortunately, we won't be seeing new Pro macs at MacWorld.
I think you are mistaken. Expect the 8-core Mac Pro to have a staring role in the January 15th SteveNote.
If you're saying apple is going to wait until NAB or "photo shows" I highly disagree. The next event after MWSF is usually in March or April. Apple can't go that long without releasing. Apple has updated these machines on normal days in the passed, but I don't see any reason for apple to release these machines between now and MWSF... and I can't imagine they go much passed MWSF.
I will be in shock if it's not January 15.
After macworld would be insanity... The mac pro specs haven't been touched since August 2006... Why wouldn't they announce it at MacWorld if it's ready? It's free Press.
Zactly.
Don't gotta apologize. It's all speculation. Apple could announce any day, any time they want. They have done it many many times in the past.
You're right, it is insanity they have gone this long. I personally don't get it. They HAVE the cpus. What else is there to wait for? Graphics cards? Motherboards? If dell has the motherboards I'm sure apple does too. All I can think of is graphics.
They had to wait for the Intel Penryn Developer Kit which was only released week before last so they could tweak Leopard for Penryn efficiency adding all the new multimedia hooks to 10.5.2 so the system software is in tune with the new Penryn, Stoakley-Seaburg hardware. They also had to wait for the supply of 3.2 GHz Harpertown Quad Core Penryns - the last to be delivered in significant quantity - to ramp up. I don't see any reason to think this next gen 2008 Mac Pro is anything but on time as soon as it could have possibly been delivered. And I also don't see how it's not going to be one of the centerpieces of Steves's message January 15th. I.E. All the pieces to the new Mac Pro will be in place by January 15, 2008. There's absolutely no reason to think it's going to be later than that.
Its going to purely be a function of how well they are yielding because, as you point out, it doesn't look like it'll be driven by competition from AMD. The mid-3 GHz range seems achievable though.
*Nods.
Quote:
Nehalem sounds like the more exciting chip because of the memory controller stuff.
Well, there's been hype about this in AMD chips for years...
The fact that its a whole new microarchitecture is the exciting part. Everybody has been clamoring for on-chip memory controllers for ages, but really I'm not sure how big a win it is in terms of performance in desktop systems. It should be a significant cost and TDP win since there are fewer parts, and it should reduce the amount of L2 cache they are obliged to include... but in terms of real-world performance due to the on-die memory controller alone, I don't think it'll live up to the hype around the issue. The new chip interconnect will be a nice addition at the MacPro level -- finally a competitor to HyperTransport... but again, on most software is it going to be a big win? The return of HyperThreading is also interesting for software that needs lots of threads, but really... with 4+ cores already are most users going to see a difference? Perhaps as software evolves over the next few years, but on the day these things ship I doubt it.
Well, I was wondering why I should wait for Nehalem when a quad or 8 way Penryn would surely do just fine. And surely, in light of current Mac Pros... Grr...the GPU is the more exciting prospect? A 9900 will seem very special after an Ati 1900 I would have thought? And in terms of 3D...Lightwave...the odd game, lots of Photoshopping, a quad or 8 core with a decent GPU would do me just fine.
Even most consumer Apple apps only seem to use 1 and a half cores half the time, heh. I think the big win is going to have to come from Software makers over the next few years.
Didn't programmers use to have to consider the chips in the Amiga? Are programmers just aiming for the 'middle ground' with their code? Are they writing lazy 'mono-thread' code..? Or do they just need to wait until Quad/Duo core reaches critical mass before they change the 'mhz' culture code?
I guess you can provide insight into that...
The GPU core tightly coupled to the CPU may be the most interesting technology. Intel has come a long way with their integrated GPUs, and I would guess that this thing is the next generation after the current X3100 and it will benefit from high speed interconnects with the CPU and the CPU's on-chip memory controller. There is no doubt that all the high-end GPU fans will slag this mercilessly just like they do the X3100 and its predecessors... but the fact remains that for the iMac and MacBook this will be a big step up. Possibly a very big step up.
Well, it seems Intel are coming round the mountain with GPUs. If they're going to exceed the x3100 then surely you'll have GPU performance in excess of a Nintendo Wii? Hey, don't knock it folks. Nintendo have bowed out the arms race and seem to be doing ok...
Quote:
What's Nehalem going to offer worth waiting a year for?
Yeah, that's my point. With Penyrn the transition to 45nm is a win, along with various architectural refinements... Nehalem is going to bring a bunch of new forward-looking stuff, but its not clear how big an improvement will be delivered on day 1.
But like many forward thinking things...we're not going to see the benefit until Quad core and maybe even 8core are mainstream? Most software seems to be only just emerging from the dark ages in terms of threading?
Looks at Photoshop. Apple demo'd funhouse as a real time filters app. Yet, did Adobe put that realtime functionality in? Nooooope.
Software makers have to get on board. Maybe it will be the 'Coco' Mac developing 'young guns' that will show what the Mac can really do now we have Leopard out.
Addendum: I just came across an Intel page discussing Nehalem, and the stuff it talks about is mostly of interest to Intel and PC makers. It is aimed at created a chip family that scales from the low end to the high end, from the low-power to the high-power. There is no discussion of specific performance enhancements or new technologies that should make a buyer wait for it. Laptops and AIO machines will benefit the most, it looks like.
A further reason for Apple's laptop sales curve to increase.
But a dinosaur like me wants a honking quad/8 core gpu monster. Hmmm.
A commendable post, Programmer. Thankyou for doing the leg work. You've certainly given me leave to buy the Mac Pro without worrying about being left behind by Nehalem. After an 8 year wait...I think Feb' might be the time to 'come home'. *Looks to the horizon.
Lemon Bon Bon.
Lemon Bon Bon.
PS. I've kinda chatted inbetween your quote. Kinda hard to see in blue...
Thanks for pointing it out, I wouldn't have noticed otherwise. I think you need to learn to use the
UBB quoting directives
.
The Amiga had multiple special purpose chips, but only one "main processor". The special purpose chips were fed sequences of commands, but these weren't the same as a fully general purpose instruction set like is found in a full fledged CPU. The first mainstream desktop machines with multiple processors started showing up in the mid-90s, although there had been many specialized machines before that (including add-on boards, super computers, etc.). Some of the early attempts provided an extra processor specifically to do graphics (the first attempts at a graphics coprocessor).
Multi-threaded programming is hard, and the tools & languages currently provided for developers to do their thing aren't very good. Frankly, you probably don't want a lot of the programmers out their writing multi-threaded programs directly because it is easy to create poorly performing, unstable multi-threaded software. This is why sophisticated system software services like OpenGL, QuickTime, CoreAnimation, CoreAudio, etc. are extremely valuable... Apple takes care of the hard parts and let's the application developers get their apps done faster and simpler. Over time the tools will improve and make the task of creating multi-threaded software easier and safer. In the meantime you have to wait for the more highly skilled developers to "do it the hard way". Only relatively recently have multi-core chips become common enough to really be worthy of the attention of most developers... previously it was only a few companies who targeted these machines.
I think you are mistaken. Expect the 8-core Mac Pro to have a staring role in the January 15th SteveNote.
And I also don't see how it's not going to be one of the centerpieces of Steves's message January 15th. I.E. All the pieces to the new Mac Pro will be in place by January 15, 2008. There's absolutely no reason to think it's going to be later than that.
AMEN BROTHER!
AMEN BROTHER!
Just in case some are still confused about the penryn launches, here's what I believe is a really nice slide of Intel's 2008 schedule:
Merry Christmas.
The Amiga had multiple special purpose chips, but only one "main processor". The special purpose chips were fed sequences of commands, but these weren't the same as a fully general purpose instruction set like is found in a full fledged CPU. The first mainstream desktop machines with multiple processors started showing up in the mid-90s, although there had been many specialized machines before that (including add-on boards, super computers, etc.). Some of the early attempts provided an extra processor specifically to do graphics (the first attempts at a graphics coprocessor).
Multi-threaded programming is hard, and the tools & languages currently provided for developers to do their thing aren't very good. Frankly, you probably don't want a lot of the programmers out their writing multi-threaded programs directly because it is easy to create poorly performing, unstable multi-threaded software. This is why sophisticated system software services like OpenGL, QuickTime, CoreAnimation, CoreAudio, etc. are extremely valuable... Apple takes care of the hard parts and let's the application developers get their apps done faster and simpler. Over time the tools will improve and make the task of creating multi-threaded software easier and safer. In the meantime you have to wait for the more highly skilled developers to "do it the hard way". Only relatively recently have multi-core chips become common enough to really be worthy of the attention of most developers... previously it was only a few companies who targeted these machines.
Noted.
I guess we're just going to have to be patient with the multi-threading thing. I would have thought with programmers working on the multi-chip 360 and the PS3 consoles that design of the chips/architecture would create a mainstream environment for threading to go mainstream...especially as they move to squeeze every drop of performance out of them. I would have thought that will help or contribute to the threading gene pool in the next 3 years as these beasts hit critical mass in their millions upon millions. The incentive to multi-thread, financially will be there? And these developers can feed the techniques back into the land of computers?
Until then, at least I can be thankful that Lightwave will take advantage of as many cores as one can throw at it (according to one Newtek guy I emailed...) Plus, I guess I can play the odd game of Eve Online, some Photoshop work...all while LWave renders in the background...and do all this using Leopard's Spaces. I guess it's not all about individual apps. Sometimes the user can just multitask...that's one way to get the most out of yeh cores in the meantime... Heh.
Lemon Bon Bon.
I think you are mistaken. Expect the 8-core Mac Pro to have a staring role in the January 15th SteveNote. I will be in shock if it's not January 15.
Prepare to be in shock. See here. Although, if true, that certainly cannot be laid at Apple's door.
I still would be extremely disappointed if there is not a least a refresh of the dual core version of Mac Pro during MWSF. The only reason for not refreshing it would be if it is slowly being phased out of their product line.
Prepare to be in shock. See here. Although, if true, that certainly cannot be laid at Apple's door.
I still would be extremely disappointed if there is not a least a refresh of the dual core version of Mac Pro during MWSF. The only reason for not refreshing it would be if it is slowly being phased out of their product line.
Those quads delayed are not Xeons, they are desktop quads and Apple doesn't use any desktop cpus from Intel.
Xeon penryns have been available since late november, most manufacturers have announced workstations and servers using those new penryn Xeons, but most will only be available late december or january.
Apple may be waiting to have enough new chips (and i5400 chipsets), and sell much of their current inventory before releasing a new Mac Pro.
Especially if the line up comes with 8 cores all the way, I really think that Apple will announce it at MWSF (8 cores starting at $2200! since those new chips are priced like the old dual-core woodcrest ones).
The question is: will they release something in between the dual-core iMac and a 8-core Mac Pro in the desktop segment?
Xeon penryns have been available since late november, most manufacturers have announced workstations and servers using those new penryn Xeons, but most will only be available late december or january.
Those are differing ideas of available. I'm sure the chip has been out in sample quantities but I haven't heard of anyone shipping any machines at all.
Apple may be waiting to have enough new chips (and i5400 chipsets), and sell much of their current inventory before releasing a new Mac Pro.
Especially if the line up comes with 8 cores all the way, I really think that Apple will announce it at MWSF (8 cores starting at $2200! since those new chips are priced like the old dual-core woodcrest ones).
The question is: will they release something in between the dual-core iMac and a 8-core Mac Pro in the desktop segment?
Personally, I hope so. Apple used to offer PowerMacs as low as $1499.
Prepare to be in shock. See here. Although, if true, that certainly cannot be laid at Apple's door.
I still would be extremely disappointed if there is not a least a refresh of the dual core version of Mac Pro during MWSF. The only reason for not refreshing it would be if it is slowly being phased out of their product line.
This is why AMD can't die out. If they do, Intel have no reason to make an effort. They only need to be better than their competition to make the sale.
Prepare to be in shock. See here. Although, if true, that certainly cannot be laid at Apple's door.
I still would be extremely disappointed if there is not a least a refresh of the dual core version of Mac Pro during MWSF. The only reason for not refreshing it would be if it is slowly being phased out of their product line.
And what do you think a refresh of dual core would look like?