By: Paul A. Clayton (paaronclayton.delete@this.gmail.com), September 27, 2013 2:23 pm
Room: Moderated Discussions
someone (someone.delete@this.somewhere.com) on September 27, 2013 12:24 pm wrote:
> bakaneko (nyan.delete@this.hyan.wan) on September 27, 2013 10:10 am wrote:
[snip]
>> That's besides the point, Intel had data that
>> VLIW sucked. Hard. They may not have been able
>> to pinpoint why, but they had data that showed
>> that VLIW designs don't compete even before
>> Itanium.
>
> Wrong. A simple comparison of the data points of high end processors all
> made in 0.18 um Al interconnect bulk CMOS shows otherwise. (SPEC CPU 2k)
>
> [Power4/1300 - 822/1202 ***SOI/Cu device, perhaps 10% faster than bulk Al 0.18 um impl.***]
> McKinley i2/1000 - 810/1431
> Athlon XP/1733 - 720/613
> Willam. P4/2000 - 681/735
> Alpha ev68b/933 - 579/724 (no official submission, scaled from 0.18 um Cu ev68b 1 GHz result)
>
> Where's the problem?
McKinley was impressive (especially relative to Merced) and comparing for the same process technology would give some insight into performance from Architecture. (The FP results can probably be ignored for an argument about the suitability of VLIW. VLIW is known to be good for classic FP workloads. I would also weakly complain that SPEC CPU 2000 allowed profile-guided optimization for base results; this would give an EPIC ISA an advantage over other ISAs.)
Unfortunately one cannot exclude secondary effects. Presumably McKinley had fewer design resources available than Willamette (or even Athlon XP if one excludes any sharing at Intel between x86 and Itanium?). McKinley was only the second implementation of the ISA (compiler maturity might also have been an issue). McKinley had higher power and chip area budgets than Athlon. McKinley was also targeted at different workloads than Athlon (for SPEC CPU2000 Int, this might give Athlon an advantage). Athlon presumably had some binning advantage (it is not clear if IBM significantly binned POWER4 by not selling parts that could not reach 1GHz).
It also seems that Itanium's design teams have been artificially constrained by the EPIC design philosophy. Not only was an out-of-order design option strongly handicapped (there are comp.arch posts indicating it was required to beat a paper tiger), but I suspect that even the relatively decoupled execution of Poulson was not readily embraced. (While the 10 year delay to Poulson may merely represent a lack of design resources, given the VLIW philosophy behind Itanium, it would not be surprising that even weakly dynamic scheduling would be viewed negatively. The abandoning of a proposed one-bundle-wide implementation might be a similar hint at a bias toward more classic VLIW/EPIC philosophy--dynamic scheduling violates the explicit nature of the parallelism, narrow issue violates the parallel nature.)
The VLIW-ish design of the Itanium ISA also places some constraints on microarchitectural diversity. x86 might be so ugly that microarchitectural cleverness flourished, and RISC might be sufficiently orthogonal to implementation that such cleverness is not impeded (albeit perhaps not encouraged as much as x86 even given equal resources). Itanium has more features that target a particular set of microarchitectures. This means that it can be more effective than RISC for those microarchitectures but also means that it can less effective for other microarchitectures.
(If Itanium had only had to compete against UltraSPARC, a declining Alpha, and a POWER3 follow-on, then the high-end RISC market might have been in greater danger. POWER4 was not just a POWER3 follow-on, so Itanium faced a more challenging market than might have been expected.)
Currently, if one were designing an ISA for the equivalent of the server/workstation market, one would presumably have less emphasis on ILP and more emphasis on the memory system and power efficiency. (One might also have more emphasis on down-scaling to performance targets below those of a traditional server/workstation processor.)
> bakaneko (nyan.delete@this.hyan.wan) on September 27, 2013 10:10 am wrote:
[snip]
>> That's besides the point, Intel had data that
>> VLIW sucked. Hard. They may not have been able
>> to pinpoint why, but they had data that showed
>> that VLIW designs don't compete even before
>> Itanium.
>
> Wrong. A simple comparison of the data points of high end processors all
> made in 0.18 um Al interconnect bulk CMOS shows otherwise. (SPEC CPU 2k)
>
> [Power4/1300 - 822/1202 ***SOI/Cu device, perhaps 10% faster than bulk Al 0.18 um impl.***]
> McKinley i2/1000 - 810/1431
> Athlon XP/1733 - 720/613
> Willam. P4/2000 - 681/735
> Alpha ev68b/933 - 579/724 (no official submission, scaled from 0.18 um Cu ev68b 1 GHz result)
>
> Where's the problem?
McKinley was impressive (especially relative to Merced) and comparing for the same process technology would give some insight into performance from Architecture. (The FP results can probably be ignored for an argument about the suitability of VLIW. VLIW is known to be good for classic FP workloads. I would also weakly complain that SPEC CPU 2000 allowed profile-guided optimization for base results; this would give an EPIC ISA an advantage over other ISAs.)
Unfortunately one cannot exclude secondary effects. Presumably McKinley had fewer design resources available than Willamette (or even Athlon XP if one excludes any sharing at Intel between x86 and Itanium?). McKinley was only the second implementation of the ISA (compiler maturity might also have been an issue). McKinley had higher power and chip area budgets than Athlon. McKinley was also targeted at different workloads than Athlon (for SPEC CPU2000 Int, this might give Athlon an advantage). Athlon presumably had some binning advantage (it is not clear if IBM significantly binned POWER4 by not selling parts that could not reach 1GHz).
It also seems that Itanium's design teams have been artificially constrained by the EPIC design philosophy. Not only was an out-of-order design option strongly handicapped (there are comp.arch posts indicating it was required to beat a paper tiger), but I suspect that even the relatively decoupled execution of Poulson was not readily embraced. (While the 10 year delay to Poulson may merely represent a lack of design resources, given the VLIW philosophy behind Itanium, it would not be surprising that even weakly dynamic scheduling would be viewed negatively. The abandoning of a proposed one-bundle-wide implementation might be a similar hint at a bias toward more classic VLIW/EPIC philosophy--dynamic scheduling violates the explicit nature of the parallelism, narrow issue violates the parallel nature.)
The VLIW-ish design of the Itanium ISA also places some constraints on microarchitectural diversity. x86 might be so ugly that microarchitectural cleverness flourished, and RISC might be sufficiently orthogonal to implementation that such cleverness is not impeded (albeit perhaps not encouraged as much as x86 even given equal resources). Itanium has more features that target a particular set of microarchitectures. This means that it can be more effective than RISC for those microarchitectures but also means that it can less effective for other microarchitectures.
(If Itanium had only had to compete against UltraSPARC, a declining Alpha, and a POWER3 follow-on, then the high-end RISC market might have been in greater danger. POWER4 was not just a POWER3 follow-on, so Itanium faced a more challenging market than might have been expected.)
Currently, if one were designing an ISA for the equivalent of the server/workstation market, one would presumably have less emphasis on ILP and more emphasis on the memory system and power efficiency. (One might also have more emphasis on down-scaling to performance targets below those of a traditional server/workstation processor.)
Topic | Posted By | Date |
---|---|---|
Charlie re: Apple and ARM | jose | 2013/09/23 03:43 AM |
Charlie re: Apple and ARM | Mark Roulo | 2013/09/23 06:38 AM |
graphics and disk matter too | RichardC | 2013/09/23 11:23 AM |
Charlie re: Apple and ARM | Jose | 2013/09/24 05:56 AM |
Previous CPU transitions | Mark Roulo | 2013/09/24 06:20 AM |
Previous CPU transitions | Ronald Maas | 2013/09/24 09:21 PM |
Charlie re: Apple and ARM | Doug S | 2013/09/23 08:16 AM |
Charlie re: Apple and ARM | Patrick Chase | 2013/09/23 08:43 AM |
Charlie re: Apple and ARM | someone | 2013/09/23 08:46 AM |
Charlie re: Apple and ARM | Patrick Chase | 2013/09/23 09:17 AM |
Charlie re: Apple and ARM | Gabriele Svelto | 2013/09/23 09:24 AM |
Charlie re: Apple and ARM | Patrick Chase | 2013/09/23 09:40 AM |
Charlie re: Apple and ARM | someone | 2013/09/23 11:42 AM |
Charlie re: Apple and ARM | Patrick Chase | 2013/09/23 05:47 PM |
Charlie re: Apple and ARM | someone | 2013/09/23 08:43 AM |
Charlie re: Apple and ARM | Doug S | 2013/09/23 09:03 AM |
Charlie re: Apple and ARM | someone | 2013/09/23 09:25 AM |
Charlie re: Apple and ARM | Doug S | 2013/09/23 09:44 AM |
Charlie re: Apple and ARM | Michael S | 2013/09/23 10:02 AM |
Charlie re: Apple and ARM | someone | 2013/09/23 11:57 AM |
Charlie re: Apple and ARM | Michael S | 2013/09/23 02:56 PM |
Charlie re: Apple and ARM | Ricardo B | 2013/09/23 11:32 PM |
Charlie re: Apple and ARM | Martin Høyer Kristiansen | 2013/09/23 12:30 PM |
Charlie re: Apple and ARM | Niels Jørgen Kruse | 2013/09/23 10:09 AM |
Charlie re: Apple and ARM | Doug S | 2013/09/23 04:09 PM |
Charlie re: Apple and ARM | Maynard Handley | 2013/09/23 11:03 AM |
Charlie re: Apple and ARM | Michael S | 2013/09/23 03:27 PM |
Charlie re: Apple and ARM | Maynard Handley | 2013/09/23 03:39 PM |
Charlie re: Apple and ARM | Doug S | 2013/09/23 04:22 PM |
Charlie re: Apple and ARM | someone | 2013/09/24 07:13 AM |
Charlie re: Apple and ARM | Doug S | 2013/09/24 09:24 AM |
Charlie re: Apple and ARM | someone | 2013/09/24 09:41 AM |
Charlie re: Apple and ARM | Doug S | 2013/09/24 04:54 PM |
The decline of Itanium | Paul A. Clayton | 2013/09/24 08:52 PM |
The decline of Itanium | Kira | 2013/09/25 05:07 AM |
The decline of Itanium | Michael S | 2013/09/25 05:15 AM |
The decline of Itanium | Kira | 2013/09/25 05:21 AM |
Does Secure64 sell hardware? | Paul A. Clayton | 2013/09/25 07:18 AM |
Does Secure64 sell hardware? | Kira | 2013/09/25 08:18 AM |
Turns out they do rx2800 now. (NT) | Kira | 2013/09/25 08:20 AM |
Thanks again. RWT has some knowledgeable posters! (NT) | Paul A. Clayton | 2013/09/25 12:38 PM |
The decline of Itanium | Doug S | 2013/09/25 08:34 AM |
The decline of Itanium | David Hess | 2013/09/25 04:10 PM |
The decline of Itanium | Doug S | 2013/09/25 07:15 PM |
The decline of Itanium | David Hess | 2013/09/27 07:11 AM |
The decline of Itanium | Doug S | 2013/09/27 04:37 PM |
The decline of Itanium | David Hess | 2013/09/28 08:43 AM |
The decline of Itanium | bakaneko | 2013/09/26 02:06 PM |
The decline of Itanium | Gabriele Svelto | 2013/09/26 02:35 PM |
The decline of Itanium | Kira | 2013/09/26 03:18 PM |
The decline of Itanium | someone | 2013/09/27 07:08 AM |
The decline of Itanium | David Hess | 2013/09/27 07:20 AM |
The decline of Itanium | someone | 2013/09/27 07:56 AM |
The decline of Itanium | Linus Torvalds | 2013/09/27 11:00 AM |
i960 | someone | 2013/09/27 12:06 PM |
i960 | Michael S | 2013/09/28 08:47 AM |
i960 | JS | 2013/09/29 01:43 AM |
The decline of Itanium | David Hess | 2013/09/28 09:00 AM |
The decline of Itanium | Michael S | 2013/09/28 09:51 AM |
The decline of Itanium | David Hess | 2013/09/28 10:59 AM |
The decline of Itanium | Michael S | 2013/09/28 11:43 AM |
The decline of Itanium | David Hess | 2013/09/28 07:53 PM |
The decline of Itanium | gallier2 | 2013/09/30 12:06 AM |
x86 MCUs | Michael S | 2013/09/30 01:13 AM |
The decline of Itanium | anon | 2013/09/27 08:52 AM |
The decline of Itanium | someone | 2013/09/27 10:29 AM |
The decline of Itanium | Kira | 2013/09/27 09:19 AM |
oops - HC 1999, not 19 (NT) | Kira | 2013/09/27 10:04 AM |
The decline of Itanium | David Hess | 2013/09/27 07:06 AM |
The decline of Itanium | someone | 2013/09/27 07:25 AM |
The decline of Itanium | Linus Torvalds | 2013/09/27 09:07 AM |
The decline of Itanium | Doug S | 2013/09/27 05:09 PM |
The decline of Itanium | Maynard Handley | 2013/09/27 06:07 PM |
The decline of Itanium | Doug S | 2013/09/27 08:12 PM |
The decline of Itanium | RichardC | 2013/09/28 05:02 AM |
Laptop Design | David Hess | 2013/09/28 09:58 AM |
Laptop Design | Brett | 2013/09/28 02:14 PM |
Laptop Design | David Hess | 2013/09/28 07:35 PM |
Laptop Design | anon | 2013/09/30 01:11 AM |
Laptop Design | Brett | 2013/09/30 05:02 PM |
Laptop Design | RichardC | 2013/09/28 04:14 PM |
Laptop Design | David Hess | 2013/09/28 07:40 PM |
Laptop Design | Michael S | 2013/09/29 02:21 AM |
The decline of Itanium | Doug S | 2013/09/28 10:23 AM |
The decline of Itanium | RichardC | 2013/09/29 04:52 AM |
PS2 | Konrad Schwarz | 2013/09/29 11:53 PM |
PS2 | none | 2013/09/30 12:19 AM |
PS2 | Doug S | 2013/09/30 10:09 AM |
PS2 | sysanon | 2013/09/30 04:09 PM |
The decline of Itanium | Megol | 2013/09/29 05:35 AM |
Apple's innovations | RichardC | 2013/09/29 06:00 AM |
The decline of Itanium | Brett | 2013/09/29 01:56 PM |
The decline of Itanium | RichardC | 2013/09/29 05:00 PM |
Apple's innovations | Brett | 2013/10/10 07:20 PM |
The decline of Itanium | RichardC | 2013/09/28 04:44 AM |
The decline of Itanium | Ricardo B | 2013/09/28 04:23 PM |
The decline of Itanium | RichardC | 2013/09/29 03:51 AM |
The decline of Itanium | Ricardo B | 2013/09/29 07:27 AM |
The decline of Itanium | RichardC | 2013/09/29 11:28 AM |
The decline of Itanium | Ricardo B | 2013/09/29 03:00 PM |
The decline of Itanium | RichardC | 2013/09/29 05:07 PM |
The decline of Itanium | anon | 2013/09/30 06:04 AM |
The decline of Intel | RichardC | 2013/09/30 06:19 AM |
The decline of Intel | Kevin G | 2013/09/30 09:53 AM |
The decline of Intel | RichardC | 2013/09/30 10:13 AM |
The decline of Intel | Kevin G | 2013/10/02 08:11 AM |
The decline of Intel | tarlinian | 2013/10/02 08:27 AM |
The decline of Intel | Kevin G | 2013/10/04 09:24 AM |
450mm and EUV insertion | David Kanter | 2013/10/04 10:24 AM |
450mm and EUV insertion | tarlinian | 2013/10/04 11:23 AM |
450mm and EUV insertion | Anonym | 2013/10/04 10:39 PM |
450mm and EUV insertion | tarlinian | 2013/10/05 09:18 AM |
450mm and EUV insertion | Anonym | 2013/10/05 11:51 AM |
450mm and EUV insertion | tarlinian | 2013/10/05 12:42 PM |
450mm and EUV insertion | Anonym | 2013/10/05 02:35 PM |
450mm and EUV insertion | tarlinian | 2013/10/05 03:21 PM |
450mm and EUV insertion | David Kanter | 2013/10/07 12:48 PM |
450mm and EUV insertion | Kevin G | 2013/10/05 04:50 AM |
The decline of Intel | Brett | 2013/09/30 05:11 PM |
The decline of Intel | Purana Archer | 2013/10/01 04:52 AM |
The decline of Intel | anon | 2013/10/01 05:27 AM |
The decline of Intel | Purana Archer | 2013/10/01 06:13 AM |
The decline of Intel | mas | 2013/10/01 03:46 PM |
The decline of Intel | Purana Archer | 2013/10/01 11:26 PM |
The decline of Intel | anon | 2013/10/02 01:05 AM |
The decline of Intel | none | 2013/10/02 01:18 AM |
The decline of Intel | Purana Archer | 2013/10/02 01:35 AM |
The decline of Intel | anon | 2013/10/02 01:57 AM |
The decline of Intel | Doug S | 2013/10/02 09:08 AM |
The decline of Intel | mas | 2013/10/02 09:40 AM |
The decline of Intel | Doug S | 2013/10/02 06:32 PM |
The decline of Intel | David Kanter | 2013/10/02 09:17 PM |
Intel vs. industry gap | David Kanter | 2013/10/02 03:17 PM |
Intel vs. industry gap | Maynard Handley | 2013/10/02 04:59 PM |
Intel vs. industry gap | tarlinian | 2013/10/02 05:13 PM |
Intel vs. industry gap | Anon | 2013/10/02 11:15 PM |
Intel vs. industry gap | tarlinian | 2013/10/03 08:01 AM |
Intel vs. industry gap | David Kanter | 2013/10/02 09:10 PM |
Intel vs. industry gap | Doug S | 2013/10/03 08:59 AM |
Intel vs. industry gap | anon | 2013/10/03 03:12 PM |
Intel vs. industry gap | Doug S | 2013/10/03 03:56 PM |
Intel vs. industry gap | anon | 2013/10/03 04:48 PM |
Intel vs. industry gap | anonymou5 | 2013/10/03 04:59 PM |
Intel vs. industry gap | mas | 2013/10/04 12:10 AM |
The decline of Intel | Klimax | 2013/10/02 02:46 AM |
The decline of Intel | anon | 2013/10/02 01:53 AM |
The decline of Intel | tarlinian | 2013/10/02 08:24 AM |
The decline of Intel | David Kanter | 2013/10/01 08:06 AM |
The decline of Intel | Purana Archer | 2013/10/01 11:09 PM |
The decline of Intel | tarlinian | 2013/10/02 07:58 AM |
The decline of Intel | David Kanter | 2013/10/02 09:45 AM |
The decline of Intel | Purana Archer | 2013/10/04 05:38 AM |
The decline of Intel | David Kanter | 2013/10/04 11:41 PM |
The decline of Intel | Kevin G | 2013/10/05 07:14 AM |
The decline of Intel | Niels Jørgen Kruse | 2013/10/05 11:49 AM |
The decline of Intel | Kevin G | 2013/10/06 07:45 AM |
The decline of Intel | Doug S | 2013/10/06 09:11 PM |
The decline of Intel | Niels Jørgen Kruse | 2013/10/07 05:14 AM |
The decline of Intel | Doug S | 2013/10/07 03:36 PM |
Tool Reuse, CAPEX Efficiency | Anonym | 2013/10/02 12:37 PM |
Tool Reuse, CAPEX Efficiency | tarlinian | 2013/10/02 02:55 PM |
capex spending | Doug S | 2013/10/01 11:06 AM |
Reducing Intel's lead with less than twice the spending?? | Paul A. Clayton | 2013/10/01 04:27 PM |
Reducing Intel's lead with less than twice the spending?? | anon | 2013/10/01 07:07 PM |
Reducing Intel's lead with less than twice the spending?? | mas | 2013/10/01 10:04 PM |
Reducing Intel's lead with less than twice the spending?? | mas | 2013/10/01 10:06 PM |
Reducing Intel's lead with less than twice the spending?? | mas | 2013/10/01 10:06 PM |
Intel fabs on 22nm | Alberto | 2013/10/01 02:23 AM |
The decline of Intel | mas | 2013/10/01 03:24 PM |
The decline of Intel | anon | 2013/09/30 05:00 PM |
The decline of Itanium | David Kanter | 2013/09/29 10:19 PM |
competitive market | RichardC | 2013/09/30 05:33 AM |
competitive market | David Kanter | 2013/09/30 07:39 AM |
competitive market | RichardC | 2013/09/30 08:08 AM |
competitive market | David Kanter | 2013/09/30 11:08 AM |
competitive market | RichardC | 2013/09/30 01:00 PM |
competitive market | Anon | 2013/10/02 11:34 PM |
competitive market | Doug S | 2013/09/30 10:13 AM |
competitive market | RichardC | 2013/09/30 10:28 AM |
The decline of Itanium | Kevin G | 2013/09/27 09:07 AM |
The decline of Itanium | Maynard Handley | 2013/09/27 10:30 AM |
The decline of Itanium | someone | 2013/09/27 11:00 AM |
The decline of Itanium | TREZA | 2013/09/27 12:50 PM |
The decline of Itanium | Megol | 2013/09/27 11:52 PM |
The decline of Itanium | Maynard Handley | 2013/09/27 04:03 PM |
The decline of Itanium | anon | 2013/09/28 02:22 AM |
The decline of Itanium | Maynard Handley | 2013/09/28 08:00 AM |
That's BS | David Kanter | 2013/09/28 08:22 AM |
The decline of Itanium | anon | 2013/09/28 04:15 PM |
The decline of Itanium | mas | 2013/09/29 08:01 AM |
The decline of Itanium | mas | 2013/09/29 08:06 AM |
The decline of Itanium | Kevin G | 2013/09/29 10:06 AM |
Apple has 2-3 CPU design teams | David Kanter | 2013/09/29 10:39 AM |
The End of Moore's Law | hobold | 2013/09/30 02:00 AM |
Lower cost to process scaling can no longer be assumed. | Mark Roulo | 2013/09/30 09:50 AM |
Lower cost to process scaling can no longer be assumed. | David Kanter | 2013/09/30 12:41 PM |
Lower cost to process scaling can no longer be assumed. | EduardoS | 2013/09/30 01:05 PM |
Shouldn't the customers have *SOME* reason to move to the new process? | Mark Roulo | 2013/09/30 02:15 PM |
Shouldn't the customers have *SOME* reason to move to the new process? | mas | 2013/09/30 07:09 PM |
Shouldn't the customers have *SOME* reason to move to the new process? | Doug S | 2013/09/30 07:16 PM |
Shouldn't the customers have *SOME* reason to move to the new process? | mas | 2013/09/30 08:05 PM |
Shouldn't the customers have *SOME* reason to move to the new process? | Doug S | 2013/10/01 11:28 AM |
Shouldn't the customers have *SOME* reason to move to the new process? | mas | 2013/10/01 03:20 PM |
Shouldn't the customers have *SOME* reason to move to the new process? | Doug S | 2013/10/01 07:51 PM |
Shouldn't the customers have *SOME* reason to move to the new process? | Exophase | 2013/10/01 12:03 PM |
Shouldn't the customers have *SOME* reason to move to the new process? | mas | 2013/10/01 03:17 PM |
Shouldn't the customers have *SOME* reason to move to the new process? | Exophase | 2013/10/01 09:18 PM |
Shouldn't the customers have *SOME* reason to move to the new process? | Doug S | 2013/10/02 09:18 AM |
Shouldn't the customers have *SOME* reason to move to the new process? | Exophase | 2013/10/02 09:28 AM |
Lower cost to process scaling can no longer be assumed. | tarlinian | 2013/09/30 06:02 PM |
Lower cost to process scaling can no longer be assumed. | David Kanter | 2013/09/30 08:20 PM |
The End of Moore's Law | Greg Gritton | 2013/10/01 08:11 AM |
The End of Moore's Law | Kevin G | 2013/10/02 09:48 AM |
The decline of Itanium | Foo_ | 2013/09/28 07:50 AM |
The decline of Itanium | Ricardo B | 2013/09/28 03:17 PM |
The decline of Itanium | Dan Fay | 2013/09/27 01:51 PM |
The decline of Itanium | Michael S | 2013/09/28 09:58 AM |
The decline of Itanium | Doug S | 2013/09/28 10:39 AM |
The decline of Itanium | Michael S | 2013/09/28 12:11 PM |
The decline of Itanium | Dan Fay | 2013/09/28 02:38 PM |
The decline of Itanium | Doug S | 2013/09/28 04:09 PM |
The decline of Itanium | Dan Fay | 2013/09/28 04:59 PM |
The decline of Itanium | mas | 2013/09/29 05:45 AM |
The decline of Itanium | none | 2013/09/29 06:10 AM |
Bay Trail die cost | mas | 2013/09/29 06:31 AM |
Bay Trail die cost | none | 2013/09/29 06:40 AM |
Bay Trail die cost | mas | 2013/09/29 07:11 AM |
Bay Trail die cost | mas | 2013/09/29 07:16 AM |
Bay Trail die cost | Doug S | 2013/09/29 10:13 AM |
Bay Trail die cost | mas | 2013/09/29 10:59 AM |
Bay Trail die cost | RichardC | 2013/10/01 05:20 AM |
The decline of Itanium | bakaneko | 2013/09/29 07:59 AM |
The decline of Itanium | mas | 2013/09/29 08:16 AM |
The decline of Itanium | mas | 2013/09/29 08:31 AM |
The decline of Itanium | bakaneko | 2013/09/29 08:48 AM |
The decline of Itanium | mas | 2013/09/29 10:12 AM |
The decline of Itanium | bakaneko | 2013/09/29 10:53 AM |
The decline of Itanium | mas | 2013/09/29 11:11 AM |
The decline of Itanium | Maynard Handley | 2013/09/29 02:15 PM |
The decline of Itanium | mas | 2013/09/29 10:28 PM |
The decline of Itanium | anon | 2013/09/30 12:26 AM |
The decline of Itanium | mas | 2013/09/30 06:20 AM |
The decline of Itanium | Doug S | 2013/09/30 07:04 PM |
The decline of Itanium | mas | 2013/09/30 07:42 PM |
The decline of Itanium | anon | 2013/09/30 10:32 PM |
The decline of Itanium | David Kanter | 2013/09/30 11:43 PM |
The decline of Itanium | anon | 2013/10/01 01:37 AM |
The decline of Itanium | David Kanter | 2013/10/01 08:17 AM |
The decline of Itanium | mas | 2013/10/01 12:54 AM |
The decline of Itanium | anon | 2013/10/01 01:39 AM |
The decline of Itanium | bakaneko | 2013/09/30 03:26 AM |
The decline of Itanium | Maynard Handley | 2013/09/29 02:08 PM |
The decline of Itanium | Maynard Handley | 2013/09/29 03:50 PM |
The decline of Itanium | mas | 2013/09/29 10:42 PM |
Semiconductor realities | David Kanter | 2013/09/30 10:30 AM |
Restricted rules for initial process use at foundries? | Paul A. Clayton | 2013/09/30 03:33 PM |
Restricted rules for initial process use at foundries? | Ricardo B | 2013/09/30 11:47 PM |
$150 7" 800p Z2580 Dell Venue 7 | mas | 2013/10/02 11:10 AM |
$150 7" 800p Z2580 Dell Venue 7 | RichardC | 2013/10/03 07:51 AM |
$150 7" 800p Z2580 Dell Venue 7 | mas | 2013/10/03 08:41 AM |
$150 7" 800p Z2580 Dell Venue 7 | RichardC | 2013/10/03 09:56 AM |
$150 7" 800p Z2580 Dell Venue 7 | Michael S | 2013/10/03 09:58 AM |
$150 7" 800p Z2580 Dell Venue 7 | RichardC | 2013/10/03 10:07 AM |
cheap would be in kindle fire | RichardC | 2013/10/03 10:12 AM |
$150 7" 800p Z2580 Dell Venue 7 | none | 2013/10/03 10:13 AM |
Samsung Galaxy Tab battery life | Michael S | 2013/10/03 01:18 PM |
Samsung Galaxy Tab battery life | none | 2013/10/03 02:17 PM |
Samsung Galaxy Tab battery life | Exophase | 2013/10/03 02:42 PM |
The decline of Itanium | none | 2013/09/29 01:15 AM |
The decline of Itanium | Doug S | 2013/09/29 10:25 AM |
The decline of Itanium | mas | 2013/09/29 11:23 AM |
Qualcomm? | David Kanter | 2013/09/29 10:45 PM |
Qualcomm? | none | 2013/09/30 12:36 AM |
Qualcomm? | Alberto | 2013/10/01 08:03 AM |
Qualcomm? | Alberto | 2013/10/01 12:03 PM |
A7 much faster at graphics than BayTrail | Thu | 2013/09/28 07:52 PM |
A7 much faster at graphics than BayTrail | Michael S | 2013/09/29 01:24 AM |
A7 much faster at graphics than BayTrail | Maynard Handley | 2013/09/29 08:41 AM |
A7 much faster at graphics than BayTrail | bakaneko | 2013/09/29 09:44 AM |
A7 much faster at graphics than BayTrail | Linus Torvalds | 2013/09/29 01:22 PM |
A7 much faster at graphics than BayTrail | none | 2013/09/29 02:37 PM |
The decline of Itanium | anoanon | 2013/09/28 03:14 AM |
The decline of Itanium | Doug S | 2013/09/28 10:44 AM |
The decline of Itanium | David Hess | 2013/09/28 08:31 AM |
The decline of Itanium | Kevin G | 2013/09/27 08:47 AM |
The decline of Itanium | David Hess | 2013/10/05 05:35 PM |
The decline of Itanium | Kevin G | 2013/10/06 07:55 AM |
The decline of Itanium | Michael S | 2013/10/06 08:13 AM |
The decline of Itanium | bakaneko | 2013/09/27 09:10 AM |
The decline of Itanium | someone | 2013/09/27 11:24 AM |
The decline of Itanium | EduardoS | 2013/09/27 12:39 PM |
The decline of Itanium | someone | 2013/09/27 01:38 PM |
The decline of Itanium | EduardoS | 2013/09/27 02:49 PM |
The decline of Itanium | someone | 2013/09/28 08:20 AM |
The decline of Itanium | EduardoS | 2013/09/28 10:05 AM |
The decline of Itanium | anon | 2013/09/27 08:22 PM |
The decline of Itanium | Maynard Handley | 2013/09/27 11:45 PM |
The decline of Itanium | anon | 2013/09/28 02:08 AM |
The decline of Itanium | EduardoS | 2013/09/28 10:08 AM |
The decline of Itanium | anon | 2013/09/28 04:17 PM |
The decline of Itanium | Michael S | 2013/09/29 02:29 AM |
The decline of Itanium | bakaneko | 2013/09/27 12:41 PM |
Difficulty of measuring performance from Architecture | Paul A. Clayton | 2013/09/27 02:23 PM |
Difficulty of measuring performance from Architecture | someone | 2013/09/27 03:46 PM |
Difficulty of measuring performance from Architecture | EduardoS | 2013/09/27 03:52 PM |
Difficulty of measuring performance from Architecture | someone | 2013/09/27 04:10 PM |
The decline of Itanium | Maynard Handley | 2013/09/27 04:09 PM |
The decline of Itanium | Michael S | 2013/09/28 10:19 AM |
why did you exclude EV7? | Michael S | 2013/09/28 10:16 AM |
why did you exclude EV7? | slacker | 2013/09/28 07:37 PM |
why did you exclude EV7? | Michael S | 2013/09/28 11:50 PM |
Wasn't Athlon XP also copper interconnect? (NT) | Paul A. Clayton | 2013/09/29 09:06 AM |
Wasn't Athlon XP also copper interconnect? | slacker | 2013/09/29 02:17 PM |
Was the SPEC CPU2000 result CU or Al? | Paul A. Clayton | 2013/09/30 04:14 PM |
Was the SPEC CPU2000 result CU or Al? | slacker | 2013/10/01 01:48 AM |
The decline of Itanium | Ricardo B | 2013/09/28 03:23 PM |
The decline of Itanium | Michael S | 2013/09/29 02:46 AM |
The decline of Itanium | Megol | 2013/09/27 10:02 AM |
The decline of Itanium | Michael S | 2013/09/28 12:31 PM |
Charlie re: Apple and ARM | Simon Farnsworth | 2013/09/25 03:06 AM |
Charlie re: Apple and ARM | anon | 2013/09/25 03:22 AM |
Charlie re: Apple and ARM | Simon Farnsworth | 2013/09/25 04:32 AM |
Charlie re: Apple and ARM | anon | 2013/09/25 04:59 AM |
Charlie re: Apple and ARM | David Kanter | 2013/09/25 12:26 PM |
Charlie re: Apple and ARM | anon | 2013/09/25 04:32 PM |
future of eDRAM | bakaneko | 2013/09/25 05:58 AM |
future of eDRAM | anon | 2013/09/25 06:43 AM |
future of eDRAM | bakaneko | 2013/09/25 08:00 AM |
future of eDRAM | anon | 2013/09/25 08:24 AM |
future of eDRAM | bakaneko | 2013/09/25 10:46 AM |
future of eDRAM | anon | 2013/09/25 04:39 PM |
future of eDRAM | bakaneko | 2013/09/26 09:51 AM |
future of eDRAM | David Kanter | 2013/09/28 09:29 AM |
future of eDRAM | bakaneko | 2013/09/27 04:23 AM |
Charlie re: Apple and ARM | Kevin G | 2013/09/25 06:18 AM |
Charlie re: Apple and ARM | anon | 2013/09/25 07:02 AM |
Charlie re: Apple and ARM | Maynard Handley | 2013/09/25 09:23 AM |
Charlie re: Apple and ARM | anon | 2013/09/25 09:59 AM |
Charlie re: Apple and ARM | Niels Jørgen Kruse | 2013/09/25 10:59 AM |
Charlie re: Apple and ARM | Maynard Handley | 2013/09/25 11:46 AM |
Charlie re: Apple and ARM | Doug S | 2013/09/25 01:15 PM |
POWER8 has 8 threads per core | Mark Roulo | 2013/09/25 03:18 PM |
Charlie re: Apple and ARM | someone | 2013/09/25 07:07 AM |
Thanks, very informative (NT) | anon | 2013/09/25 07:11 AM |
Keep in mind IBM has eDRAM elsewhere than POWER (NT) | anon | 2013/09/25 10:03 AM |
Charlie re: Apple and ARM | RichardC | 2013/09/25 06:12 AM |
It isn't just memory controllers | Mark Roulo | 2013/09/25 08:09 AM |
Charlie re: Apple and ARM | Foo_ | 2013/09/23 11:52 PM |
Charlie re: Apple and ARM | Drazick | 2013/09/23 09:29 AM |
Charlie re: Apple and ARM | Maynard Handley | 2013/09/23 10:55 AM |
Charlie re: Apple and ARM | Drazick | 2013/09/23 11:00 AM |
Charlie re: Apple and ARM | Michael S | 2013/09/23 03:01 PM |
Charlie re: Apple and ARM | Doug S | 2013/09/23 04:31 PM |
Charlie re: Apple and ARM | Maynard Handley | 2013/09/23 06:34 PM |
Charlie re: Apple and ARM | Alberto | 2013/09/24 12:11 PM |
Charlie re: Apple and ARM | Wilco | 2013/09/24 05:17 PM |
Charlie re: Apple and ARM | Doug S | 2013/09/24 07:44 PM |
Charlie re: Apple and ARM | anon | 2013/09/25 12:56 AM |
Charlie re: Apple and ARM | none | 2013/09/25 01:50 AM |
Charlie re: Apple and ARM | anon | 2013/09/25 02:06 AM |
Charlie re: Apple and ARM | Wilco | 2013/09/25 02:14 AM |
Charlie re: Apple and ARM | anon | 2013/09/25 02:28 AM |
Charlie re: Apple and ARM | Michael S | 2013/09/25 03:24 AM |
Charlie re: Apple and ARM | none | 2013/09/25 03:55 AM |
Charlie re: Apple and ARM | EduardoS | 2013/09/25 01:07 PM |
Charlie re: Apple and ARM | Doug S | 2013/09/25 09:01 AM |
Charlie re: Apple and ARM | Alberto | 2013/09/25 12:12 PM |
Charlie re: Apple and ARM | Maynard Handley | 2013/09/25 01:23 PM |
Charlie re: Apple and ARM | Wilco | 2013/09/25 01:45 AM |
Charlie re: Apple and ARM | Linus Torvalds | 2013/09/25 04:49 PM |
Charlie re: Apple and ARM | Michael S | 2013/09/26 09:52 AM |
Charlie re: Apple and ARM | Doug S | 2013/09/26 10:51 AM |
Animated GIF seems slow on iPads | Mark Roulo | 2013/09/26 12:04 PM |
Animated GIF seems slow on iPads | Doug S | 2013/09/26 01:07 PM |
Animated GIF seems slow on iPads | Mark Roulo | 2013/09/26 02:06 PM |
Animated GIF seems slow on iPads | Doug S | 2013/09/26 05:21 PM |
Animated GIF seems slow on iPads | rwessel | 2013/09/26 05:44 PM |
Animated GIF seems slow on iPads | sysanon | 2013/09/27 03:33 PM |
Animated GIF seems slow on iPads | Doug S | 2013/09/27 05:29 PM |
Animated GIF seems slow on iPads | sysanon | 2013/09/27 07:36 PM |
Animated GIF seems slow on iPads | Doug S | 2013/09/27 08:07 PM |
Animated GIF seems slow on iPads | anonymou5 | 2013/09/27 11:58 PM |
Animated GIF seems slow on iPads | J.Random Webmasta | 2013/09/28 12:11 AM |
Slow with Core i7 920 | Jouni Osmala | 2013/09/26 10:25 PM |
Animated GIF seems slow on iPads | NoSpammer | 2013/09/27 12:13 AM |
Charlie re: Apple and ARM | Maynard Handley | 2013/09/26 12:18 PM |
Charlie re: Apple and ARM | Doug S | 2013/09/26 01:19 PM |
Charlie re: Apple and ARM | Maynard Handley | 2013/09/26 01:35 PM |
Charlie re: Apple and ARM | John Poole | 2013/09/26 02:11 PM |
Charlie re: Apple and ARM | Doug S | 2013/09/26 05:31 PM |
Charlie re: Apple and ARM | John Poole | 2013/09/27 10:02 PM |
Firefox PDF reader (re: Charlie re: Apple and ARM) | David W | 2013/09/27 12:47 AM |
Firefox PDF reader (re: Charlie re: Apple and ARM) | David Kanter | 2013/09/28 09:09 AM |
Firefox PDF reader (re: Charlie re: Apple and ARM) | David Hess | 2013/09/28 09:21 AM |
Firefox PDF reader (re: Charlie re: Apple and ARM) | Michael S | 2013/09/28 10:00 AM |
Firefox PDF reader (re: Charlie re: Apple and ARM) | David Hess | 2013/09/28 10:27 AM |
Firefox PDF reader (re: Charlie re: Apple and ARM) | bakaneko | 2013/09/28 11:11 AM |
Firefox PDF reader (re: Charlie re: Apple and ARM) | Michael S | 2013/09/28 11:50 AM |
Firefox PDF reader (re: Charlie re: Apple and ARM) | EduardoS | 2013/09/28 12:50 PM |
Firefox PDF reader (re: Charlie re: Apple and ARM) | Michael S | 2013/09/28 01:05 PM |
Firefox PDF reader (re: Charlie re: Apple and ARM) | Doug S | 2013/09/28 04:15 PM |
Firefox PDF reader (re: Charlie re: Apple and ARM) | David Hess | 2013/09/28 07:03 PM |
Firefox PDF reader (re: Charlie re: Apple and ARM) | Gabriele Svelto | 2013/09/30 03:23 AM |
Firefox PDF reader (re: Charlie re: Apple and ARM) | Jukka Larja | 2013/09/30 06:23 AM |
Firefox PDF reader (re: Charlie re: Apple and ARM) | Doug S | 2013/09/30 07:19 PM |
Firefox PDF reader (re: Charlie re: Apple and ARM) | Jukka Larja | 2013/10/01 03:55 AM |
Firefox PDF reader (re: Charlie re: Apple and ARM) | Rob Thorpe | 2013/10/01 07:26 AM |
Firefox PDF reader (re: Charlie re: Apple and ARM) | Michael S | 2013/10/01 12:53 PM |
Adobe Acrobat reader start up time | Michael S | 2013/10/02 12:19 AM |
Adobe Acrobat reader start up time | bdcrazy | 2013/10/11 05:28 AM |
Firefox PDF reader - am I the only person who likes the default | Rob Thorpe | 2013/10/01 07:14 AM |
Firefox PDF reader - am I the only person who likes the default | j | 2013/10/01 10:12 AM |
There are two of us (or three) | Mark Roulo | 2013/10/01 12:15 PM |
Firefox PDF reader - am I the only person who likes the default | Rob Thorpe | 2013/10/01 03:05 PM |
Firefox PDF reader - am I the only person who likes the default | Symmetry | 2013/10/02 11:51 AM |
Firefox PDF reader - am I the only person who likes the default | Doug S | 2013/10/02 06:44 PM |
Firefox PDF reader - am I the only person who likes the default | rwessel | 2013/10/02 10:21 PM |
Firefox PDF reader - am I the only person who likes the default | Clemens Ladisch | 2013/10/02 11:20 PM |
Firefox PDF reader - am I the only person who likes the default | rwessel | 2013/10/03 12:12 AM |
Firefox PDF reader - am I the only person who likes the default | Symmetry | 2013/10/03 05:19 AM |
Firefox PDF reader - am I the only person who likes the default | Gabriele Svelto | 2013/10/03 01:05 AM |
Firefox PDF reader - am I the only person who likes the default | Doug S | 2013/10/03 09:15 AM |
Charlie re: Apple and ARM | John Poole | 2013/09/26 01:59 PM |
Charlie re: Apple and ARM | Maynard Handley | 2013/09/26 02:53 PM |
Charlie re: Apple and ARM | John Poole | 2013/10/01 09:55 PM |
Charlie re: Apple and ARM | Linus Torvalds | 2013/09/26 07:15 PM |
Charlie re: Apple and ARM | John Poole | 2013/10/01 09:45 PM |
Charlie re: Apple and ARM | Doug S | 2013/10/02 09:14 AM |
Charlie re: Apple and ARM | John Poole | 2013/10/02 09:03 PM |
Charlie re: Apple and ARM | anon | 2013/10/02 11:00 PM |
Charlie re: Apple and ARM | Doug S | 2013/10/03 09:08 AM |
Charlie re: Apple and ARM | Alberto | 2013/09/25 12:50 PM |
Charlie re: Apple and ARM | Ronald Maas | 2013/09/24 09:39 PM |