>Your whole argument is why Intel developed the Itanium. The idea of producing a simpler CPU that is thermally >more efficent is a novel one, but time and again we find that you can't erase the last 15 years of CPU innovation
Itanium still has instruction fusing (i.e. three instructions fused into a single instruction issue) and extensive HW branch support.
Note that this dual issue PPE core is a 21 stage pipeline(similar to PIV Northwood), while AMD's K8 is a 12 stage integer and 17 stage FP combo. PPE is not PPC 7447A nor it's PPC 750FX.
"you either get an OOO processor running at 2GHz with three or four issues pathways (three has been the rule under x86)"
Not quite since a K8's macro-op instruction (fix length) is fused with two instructions (one of the instructions must be an address type instruction). K8 issues three macro-op instructions. Currently, dual core K8 clocks at 2.6Ghz with Opteron SE not 2Ghz.
Note the number of processors in blue gene i.e. useless in the desktop PC market.
Refer to http://www.barefeats.com/macvpc.html
for some real world benchmarks in the desktop PC market.
>Unfortunately, the optimization units to deal
>with x86 inefficiencies end up costing nearly
>as many transistors as the units that actually >do the work
Decoder part in K8 takes about 5 percent of the total transistor budget.
According to http://chip-architect.com the Opteron(C0) has following floating point units
1X SSE2/X87 MUL,
1X SSE2/X87 ADD,
2X 64bit SSE/MMX ADD,
1X 64bit SSE/MMX MUL,
@Nicholas Blachford;
In reference to http://www.blachford.info/computer/Cells/Cell4.htm l
"Multicore processors are coming to the x86 world soon from both Intel and AMD
[MultiCore], but high speed x86 CPUs typically have high power requirements. In
order to have 2 Opterons on a single core AMD have had to reduce their clock rate in
order to keep them from requiring over a hundred watts,"
Refer to http://www.theinquirer.net/?article=20420
AMD TDPmax = 95Watts with Dual core @2400Mhz. The Mhz target is in line with the
current Opteron x50 @2400Mhz.
"Changes are not unheard of in x86 land but neither Intel or AMD appear to be
planning a change even nearly radical enough to catch up. That said Intel recently
gained access to many of Nvidia's patents [Intel+Nvidia] and are talking about
having dozens of cores per chip so who knows what Santa Clara are brewing" -
Nicholas Blachford
Refer to
http://news.cens.com/php/getnews.php?file=/news/20 04/11/30/20041130013.htm&d
"Graphics-chip supplier Nvida Corp. and microprocessor giant Intel Corp. recently
entered into alliance while ATI Technologies Inc. and Advanced Micro Devices struck
a similar deal." - China Economic News Service/The Taiwan Economic News.
"To date the PC has defeated everything in it's path [PCShare]. No competitor, no
matter how good has even got close to replacing it. If the Cell is placed into
desktop computers it may be another victim of the PC. However, I think for a number
of reasons that the Cell is not only the biggest threat the PC has ever faced, but
also one which might actually have the capacity to defeat it". - NB
That's more to a PC (refering to Microsoft PC) than just CPU (or MPU)i.e. one must factor in XNA
initiative's need for VPU/GPU/GpGPU (general-purpose GPU(1)) partnership.
Side notes.
1. http://www.cs.sunysb.edu/~vislab/projects/gpgpu/
Slashdot Mirror
Refer to
1. Laptop form factors i.e. mobile/compact PCs.
2. Thinkpad laptop's during BIOS screen i.e. hides BIOS text details.
>Your whole argument is why Intel developed the Itanium. The idea of producing a simpler CPU that is thermally >more efficent is a novel one, but time and again we find that you can't erase the last 15 years of CPU innovation Itanium still has instruction fusing (i.e. three instructions fused into a single instruction issue) and extensive HW branch support.
This analysis is incorrect...
Note that this dual issue PPE core is a 21 stage pipeline(similar to PIV Northwood), while AMD's K8 is a 12 stage integer and 17 stage FP combo. PPE is not PPC 7447A nor it's PPC 750FX.
"you either get an OOO processor running at 2GHz with three or four issues pathways (three has been the rule under x86)"
Not quite since a K8's macro-op instruction (fix length) is fused with two instructions (one of the instructions must be an address type instruction). K8 issues three macro-op instructions. Currently, dual core K8 clocks at 2.6Ghz with Opteron SE not 2Ghz.
Note the number of processors in blue gene i.e. useless in the desktop PC market. Refer to http://www.barefeats.com/macvpc.html for some real world benchmarks in the desktop PC market.
>Unfortunately, the optimization units to deal >with x86 inefficiencies end up costing nearly >as many transistors as the units that actually >do the work Decoder part in K8 takes about 5 percent of the total transistor budget.
Note that Cell's PPE (control CPU) is dual issue in-order processor. It's nothing like PowerPC 970.
Around Q1 2005, PPC (includes embedded class processors) unit sales per year is about 60 million.
0 1_05.cfm
Intel reaches ~200 million units(laptop/desktop/server class processors) per year.
Refer to www.tundra.com/NewsRoom/PressReleases/2005/pr_03_
Issues with Linpack can be gathered from http://www.netlib.org/lapack/
According to http://chip-architect.com the Opteron(C0) has following floating point units 1X SSE2/X87 MUL, 1X SSE2/X87 ADD, 2X 64bit SSE/MMX ADD, 1X 64bit SSE/MMX MUL,
Refer to http://www4.tomshardware.com/howto/20020531/window s_gaming-05.html
for the opposite view i.e. WineX vs Windows 2K.
Refer to http://money.cnn.com/2004/05/26/commentary/game_ov er/column_gaming/
@Nicholas Blachford; In reference to http://www.blachford.info/computer/Cells/Cell4.htm l
"Multicore processors are coming to the x86 world soon from both Intel and AMD
[MultiCore], but high speed x86 CPUs typically have high power requirements. In
order to have 2 Opterons on a single core AMD have had to reduce their clock rate in
order to keep them from requiring over a hundred watts,"
Refer to http://www.theinquirer.net/?article=20420
AMD TDPmax = 95Watts with Dual core @2400Mhz. The Mhz target is in line with the
current Opteron x50 @2400Mhz.
"Changes are not unheard of in x86 land but neither Intel or AMD appear to be
planning a change even nearly radical enough to catch up. That said Intel recently
gained access to many of Nvidia's patents [Intel+Nvidia] and are talking about
having dozens of cores per chip so who knows what Santa Clara are brewing" -
Nicholas Blachford
Refer to
http://news.cens.com/php/getnews.php?file=/news/20 04/11/30/20041130013.htm&d
"Graphics-chip supplier Nvida Corp. and microprocessor giant Intel Corp. recently
entered into alliance while ATI Technologies Inc. and Advanced Micro Devices struck
a similar deal." - China Economic News Service/The Taiwan Economic News.
"To date the PC has defeated everything in it's path [PCShare]. No competitor, no
matter how good has even got close to replacing it. If the Cell is placed into
desktop computers it may be another victim of the PC. However, I think for a number
of reasons that the Cell is not only the biggest threat the PC has ever faced, but
also one which might actually have the capacity to defeat it". - NB
That's more to a PC (refering to Microsoft PC) than just CPU (or MPU)i.e. one must factor in XNA
initiative's need for VPU/GPU/GpGPU (general-purpose GPU(1)) partnership.
Side notes.
1. http://www.cs.sunysb.edu/~vislab/projects/gpgpu/