Amiga.org
The "Not Quite Amiga but still computer related category" => Alternative Operating Systems => Topic started by: mingle on February 02, 2008, 12:34:46 AM
-
Hi,
I understand that since the old PPC were RISC chips, they outperformed equivalent clockspeed x86 CPUs...
Is this true?
Would a 1GHz G4 PPC be the same as, or faster than a 1GHz PIII?
How do each of the PPC processor generations (G3, G4, G5) correspond to the rival x86 families (PII, PIII, PIV, etc)?
Cheers,
Mike.
-
Would a 1GHz G4 PPC be the same as, or faster than a 1GHz PIII?
You can't compare two different processors by their clock speed. It's like an engine; revolutions per minute doesn't say anything about how fast the car moves.
-
You cant even compare different genrations of the same processor family by MHZ. MHZ is the most pointless indicator of chip performace of chips thers is. Unless you are comparing chips of eexactly the same architecture. Obivously 50mhz 030 is faster than a 25mhz 030. But you cant compare it against anything else by the MHZ
-
Okay, I understand that MHz isn't a good benchmark...
What I'm getting at is how they compare, ie: MIPs/ Dhrystones/ Real World performance...
If I have a 1.8GHz G5 Mac and a 2.0GHz P4, running the same OS (Say a flavour of Linux), which would yield better performance?
-
If I have a 1.8GHz G5 Mac and a 2.0GHz P4, running the same OS (Say a flavour of Linux), which would yield better performance?
Depends on what you do, but G5 should beat P4 at the same clockrate at many tasks.
The thing is P4 is stoneage x86 technology these days. Take modern core2 cpu .. it runs 1333 fsb (and new ones 1600) and it runs circles around any ppc. Runs cooler, too.
-
mingle wrote:
Okay, I understand that MHz isn't a good benchmark...
What I'm getting at is how they compare, ie: MIPs/ Dhrystones/ Real World performance...
If I have a 1.8GHz G5 Mac and a 2.0GHz P4, running the same OS (Say a flavour of Linux), which would yield better performance?
The P4 is pretty lame... The G5 will be at least as fast if not quite a bit faster... If you had an Athlon64 or a CoreDuo then they would piss all over the G5 :-)
-
Piru wrote:
If I have a 1.8GHz G5 Mac and a 2.0GHz P4, running the same OS (Say a flavour of Linux), which would yield better performance?
Depends on what you do, but G5 should beat P4 at the same clockrate at many tasks.
The thing is P4 is stoneage x86 technology these days. Take modern core2 cpu .. it runs 1333 fsb (and new ones 1600) and it runs circles around any ppc. Runs cooler, too.
But throw it against an UltraSPARC T2, and it will be crushed in both heat and performance, despite the T2 being half the clock speed.
-
downix wrote:
But throw it against an UltraSPARC T2, and it will be crushed in both heat and performance, despite the T2 being half the clock speed.
The T2 isn't quite in the "consumer" class. Sure, it's the meanest thing out there, for now. But since the cheapest box that comes with it is >$10k, it's really unfair to compare it with processors that cost <$200.
-
adolescent wrote:
downix wrote:
But throw it against an UltraSPARC T2, and it will be crushed in both heat and performance, despite the T2 being half the clock speed.
The T2 isn't quite in the "consumer" class. Sure, it's the meanest thing out there, for now. But since the cheapest box that comes with it is >$10k, it's really unfair to compare it with processors that cost <$200.
Oh Sure, tie my hands for comparison.... *tee hee*
and where do you fine a core2 for under $200? Lowest price I can find locally is $350!
-
and where do you fine a core2 for under $200? Lowest price I can find locally is $350!
Pricewatch is your friend (http://www.stalliontek.com/ProductDetails.asp?ProductCode=CPC2D%2DE6600OEM).
Dammy
-
Although what was said already is good and to the point, I'd like to add (for the original poster or others in his predisposition) that x86 CPUs after and including the PentiumPro (as far Intel, and K5 I think for AMD, if not then K6) are all RISC internally. They convert x86 assembly/machine code into micro-ops (operations) which execute on a RISC core. So they're not really that far removed from what PPC chips, as far as high-level architecture. Of course the devil's in the details...
-
Sparc (http://en.wikipedia.org/wiki/SPARC) on Wikipedia.
As a result of SPARC International, the SPARC architecture is fully open and non-proprietary.
Implementations of the SPARC architecture were initially designed and used for Sun's Sun-4 workstation and server systems, replacing their earlier Sun-3 systems based on the Motorola 68000 family of processors.
Is there any chance of the sparc (any version of it) running Amiga OS as well???
-
Is there any chance of the sparc (any version of it) running Amiga OS
no
-
$207 for a year old revision, riiiight.
dammy wrote:
and where do you fine a core2 for under $200? Lowest price I can find locally is $350!
Pricewatch is your friend (http://www.stalliontek.com/ProductDetails.asp?ProductCode=CPC2D%2DE6600OEM).
Dammy
-
pyrre wrote:
Sparc (http://en.wikipedia.org/wiki/SPARC) on Wikipedia.
As a result of SPARC International, the SPARC architecture is fully open and non-proprietary.
Implementations of the SPARC architecture were initially designed and used for Sun's Sun-4 workstation and server systems, replacing their earlier Sun-3 systems based on the Motorola 68000 family of processors.
Is there any chance of the sparc (any version of it) running Amiga OS as well???
Would be an easy job (I've gotten parts of AROS running w/o any real programming skill) especially if one were to use the migration toolchain Sun developed back when they introduced the Sun-4, especially as the Amiga's OS was built on the Sun-3 to begin with!
-
@downix
Would be an easy job especially if one were to use the migration toolchain Sun developed back when they introduced the Sun-4
You obviously would need the source code to do that. It won't happen.
-
The P4 is pretty lame... The G5 will be at least as fast if not quite a bit faster... If you had an Athlon64 or a CoreDuo then they would piss all over the G5
CoreDuo yes its faster but...
Athlon64??? maybe you mean the AthlonX2 ??
because my brother has an Athlon64FX@3200Mhz and I have a DUAL1.5Ghz G4 and I am 2 times faster in almost everything....
ofcourse athlon is running WinXP64 and I'm running OSX, but I don't think that an OS can make such a difference anyway...
EDIT: I think in INTEGER mulithreaded operations we must be at about the same speed, but in video encodings like H.264 and renderings I'm about 2x faster..
maybe thats because of 4 altivec units that my machine has..(2 altivecs per cpu)
-
Would be an easy job (I've gotten parts of AROS running w/o any real programming skill) especially if one were to use the migration toolchain Sun developed back when they introduced the Sun-4, especially as the Amiga's OS was built on the Sun-3 to begin with!
Cool it would be insane running amiga os on a sun platform, however... expensive...
-
$207 for a year old revision, riiiight.Quote:
When I linked that page, it was under $200 with shipping. Prices go up and down a couple of dollars, but it was under $200 and you never specify how old the series had to be.
Dammy
-
EDIT: I think in INTEGER mulithreaded operations we must be at about the same speed, but in video encodings like H.264 and renderings I'm about 2x faster..
maybe thats because of 4 altivec units that my machine has..(2 altivecs per cpu)
What video cards are you using in the two machines?
Dammy
-
@dammy
What video cards are you using in the two machines?
What gfxcard has to do with H.264 encoding or renderings?
-
What video cards are you using in the two machines?
Dammy
Actually I use an Ati X850XT and my brother a miserable Nvidia 5200FX
but renderings have nothing to do with graphic cards!!!
well in preview mode or realtime FXs preview like some from imovie yes. but exporting to dvix,H.264 or any other format has NOTHING to do with GFX cards..
-
@dammy
Quote:
What video cards are you using in the two machines?
What gfxcard has to do with H.264 encoding or renderings?
Oh SOOOOOOOOO sorry that I ASKED a question. :roll:
Dammy
-
Actually I use an Ati X850XT and my brother a miserable Nvidia 5200FXActually I use an Ati X850XT and my brother a miserable Nvidia 5200FX
but renderings have nothing to do with graphic cards!!!
well in preview mode or realtime FXs preview like some from imovie yes. but exporting to dvix,H.264 or any other format has NOTHING to do with GFX cards..
but renderings have nothing to do with graphic cards!!!
well in preview mode or realtime FXs preview like some from imovie yes. but exporting to dvix,H.264 or any other format has NOTHING to do with GFX cards..
Ah, thanks.
Dammy
-
@dammy
Oh SOOOOOOOOO sorry that I ASKED a question.
Oh please.
-
@StormLord
I think in INTEGER mulithreaded operations we must be at about the same speed, but in video encodings like H.264 and renderings I'm about 2x faster..
maybe thats because of 4 altivec units that my machine has..(2 altivecs per cpu)
That's it, most likely. AltiVec (especially in 7447/8) is still better than vector units in x86 class machines. Dual G4 is pretty formidable with sw that has good altivec support. The more CPU bound the task is, the better G4 and altivec does (that is G4 doesn't have ultrafast frontside bus and altivec can easily get starved, x86 are tons faster with highly memory intensive tasks).
-
Well, I have both a Athlon64 3200 (Which I think is 2.2Ghz) and a PPC 7447A (ie a 1.5Ghz G4 in my old Mac)... I can confirm that the G4 is about as half as fast in every operation as the Athlon64. Which is what one would expect. Pretty much all my tests are based on using Logic Pro...
If someone has any cross platform benchmarking tools they would like me to run, I would be happy to and will publish the results. (I also have a 3Ghz P4 as well)
The 2.4Ghz Core2Duo in my MacBook Pro... out classes all of those other machines!
-
Piru wrote:
@downix
Would be an easy job especially if one were to use the migration toolchain Sun developed back when they introduced the Sun-4
You obviously would need the source code to do that. It won't happen.
Well, quite obviously. AInc or Hyperion own that, and neither one seem inclined for such a move. Their choice, of course.
-
pyrre wrote:
Would be an easy job (I've gotten parts of AROS running w/o any real programming skill) especially if one were to use the migration toolchain Sun developed back when they introduced the Sun-4, especially as the Amiga's OS was built on the Sun-3 to begin with!
Cool it would be insane running amiga os on a sun platform, however... expensive...
How so? SPARC's are licenseable, and have multiple vendors. Don't like Suns pricing, go to TI, Fuji, Geisler, even Freescale has a license to produce them.
-
Piru wrote:
@StormLord
I think in INTEGER mulithreaded operations we must be at about the same speed, but in video encodings like H.264 and renderings I'm about 2x faster..
maybe thats because of 4 altivec units that my machine has..(2 altivecs per cpu)
That's it, most likely. AltiVec (especially in 7447/8) is still better than vector units in x86 class machines. Dual G4 is pretty formidable with sw that has good altivec support. The more CPU bound the task is, the better G4 and altivec does (that is G4 doesn't have ultrafast frontside bus and altivec can easily get starved, x86 are tons faster with highly memory intensive tasks).
I always found the PowerPC's FSB always the achilees heel of the design. It always came down to CPU-bound tasks vs non-CPU-bound tasks for performance. If I were PPC's controlling execs a few years back, I'd have worked with AMD to bring Hypertransport to the PowerPC, with an integrated DDR RAM controller, similar to how the early Opterons worked. but, opportunities missed leaves new ones ahead, right?
-
Piru wrote:
Is there any chance of the sparc (any version of it) running Amiga OS
no
UAE ! ;-)
-
JJ wrote:
You cant even compare different genrations of the same processor family by MHZ. MHZ is the most pointless indicator of chip performace of chips thers is. Unless you are comparing chips of eexactly the same architecture. Obivously 50mhz 030 is faster than a 25mhz 030. But you cant compare it against anything else by the MHZ
Although saying that, I do recall a 'rule of thumb' that a PPC of a certain clock could perform around 2x the equivalently clocked contemporary Pentium. I guess that'd be the PIII cf. the G3 though...
-
Piru wrote:
@dammy
What video cards are you using in the two machines?
What gfxcard has to do with H.264 encoding or renderings?
I have an H.264 decoder in my video system here.
*edit, should specify*
It's a video decoder card I'm working on, in FPGA, plugged into one of my PCI ports, based on the newly open sourced MIT code.
-
A decoder is not an encoder..
Altough there are hardware based h.264 encoders. And with newer gfx cards it's possible to use the raw power of the graphics chip for other purposes then generating images.
-
Agafaster wrote:
JJ wrote:
You cant even compare different genrations of the same processor family by MHZ. MHZ is the most pointless indicator of chip performace of chips thers is. Unless you are comparing chips of eexactly the same architecture. Obivously 50mhz 030 is faster than a 25mhz 030. But you cant compare it against anything else by the MHZ
Although saying that, I do recall a 'rule of thumb' that a PPC of a certain clock could perform around 2x the equivalently clocked contemporary Pentium. I guess that'd be the PIII cf. the G3 though...
No, that would be against the Pentium4 which had very long (22 stage?) pipelines, so that very high clock speeds could be achieved, at the expense of work that could be done per clock cycle. It was a strategy based on the idea that transistor switching speeds would increase dramatically in a short space of time... this did not happen and they have currently topped out at around ~3Ghz.
AMD and Motorola opted for shorter pipelines, which resulted in lower clock speeds but more work gets done per cycle... With the Core2 architecture Intel have adopted the same approach, and brought with them all the good stuff from the P4 (ie great branch predictors and macroop fusion, etc)... (and borrowed all the good ideas of the Athlon64 and the PIII too)...
-
bloodline wrote:
Agafaster wrote:
JJ wrote:
You cant even compare different genrations of the same processor family by MHZ. MHZ is the most pointless indicator of chip performace of chips thers is. Unless you are comparing chips of eexactly the same architecture. Obivously 50mhz 030 is faster than a 25mhz 030. But you cant compare it against anything else by the MHZ
Although saying that, I do recall a 'rule of thumb' that a PPC of a certain clock could perform around 2x the equivalently clocked contemporary Pentium. I guess that'd be the PIII cf. the G3 though...
No, that would be against the Pentium4 which had very long (22 stage?) pipelines, so that very high clock speeds could be achieved, at the expense of work that could be done per clock cycle. It was a strategy based on the idea that transistor switching speeds would increase dramatically in a short space of time... this did not happen and they have currently topped out at around ~3Ghz.
AMD and Motorola opted for shorter pipelines, which resulted in lower clock speeds but more work gets done per cycle... With the Core2 architecture Intel have adopted the same approach, and brought with them all the good stuff from the P4 (ie great branch predictors and macroop fusion, etc)... (and borrowed all the good ideas of the Athlon64 and the PIII too)...
With the irony being that Intel's next-gen chips are going back to the classic P4 methodology...
-
downix wrote:
bloodline wrote:
Agafaster wrote:
JJ wrote:
You cant even compare different genrations of the same processor family by MHZ. MHZ is the most pointless indicator of chip performace of chips thers is. Unless you are comparing chips of eexactly the same architecture. Obivously 50mhz 030 is faster than a 25mhz 030. But you cant compare it against anything else by the MHZ
Although saying that, I do recall a 'rule of thumb' that a PPC of a certain clock could perform around 2x the equivalently clocked contemporary Pentium. I guess that'd be the PIII cf. the G3 though...
No, that would be against the Pentium4 which had very long (22 stage?) pipelines, so that very high clock speeds could be achieved, at the expense of work that could be done per clock cycle. It was a strategy based on the idea that transistor switching speeds would increase dramatically in a short space of time... this did not happen and they have currently topped out at around ~3Ghz.
AMD and Motorola opted for shorter pipelines, which resulted in lower clock speeds but more work gets done per cycle... With the Core2 architecture Intel have adopted the same approach, and brought with them all the good stuff from the P4 (ie great branch predictors and macroop fusion, etc)... (and borrowed all the good ideas of the Athlon64 and the PIII too)...
With the irony being that Intel's next-gen chips are going back to the classic P4 methodology...
Well... err... only as far as I can see... The silverthorn, with its simple in order pipeline... but that is with good reason, i.e. to get a small die and low power consumption so it can compete with the ARM...
-
downix wrote:
bloodline wrote:
Agafaster wrote:
JJ wrote:
You cant even compare different genrations of the same processor family by MHZ. MHZ is the most pointless indicator of chip performace of chips thers is. Unless you are comparing chips of eexactly the same architecture. Obivously 50mhz 030 is faster than a 25mhz 030. But you cant compare it against anything else by the MHZ
Although saying that, I do recall a 'rule of thumb' that a PPC of a certain clock could perform around 2x the equivalently clocked contemporary Pentium. I guess that'd be the PIII cf. the G3 though...
No, that would be against the Pentium4 which had very long (22 stage?) pipelines, so that very high clock speeds could be achieved, at the expense of work that could be done per clock cycle. It was a strategy based on the idea that transistor switching speeds would increase dramatically in a short space of time... this did not happen and they have currently topped out at around ~3Ghz.
AMD and Motorola opted for shorter pipelines, which resulted in lower clock speeds but more work gets done per cycle... With the Core2 architecture Intel have adopted the same approach, and brought with them all the good stuff from the P4 (ie great branch predictors and macroop fusion, etc)... (and borrowed all the good ideas of the Athlon64 and the PIII too)...
With the irony being that Intel's next-gen chips are going back to the classic P4 methodology...
...but maybe for different reasons - other than the cynical attempt to Ramp up the Clock speed for Marketing to Dullards(tm) purposes
-
bloodline wrote:
Well... err... only as far as I can see... The silverthorn, with its simple in order pipeline... but that is with good reason, i.e. to get a small die and low power consumption so it can compete with the ARM...
Not the Silverthorn, their next-gen Xeons seem to be abandoning the Core-like system and going backwards to P4-style. I think a bad move from all angles.
-
If someone has any cross platform benchmarking tools they would like me to run, I would be happy to and will publish the results. (I also have a 3Ghz P4 as well
even crossplatform benchmarking can't show the real difference and that because most of them are optimised and builted for one type of machines and just crossed compiled to the other, instead of rewriting the program from scratch to take care different architectures.. also different architectures can have very different results in different tests
Example: try to run distributed net OGR to that single 1.5Ghz G4 and OGR on that 3200AMD, both cores are written from scratch for the specific CPUs , results:
G4 will be about 3-4x faster than the athlon, but that is not true in everyday tasks performance
-
StormLord wrote:
If someone has any cross platform benchmarking tools they would like me to run, I would be happy to and will publish the results. (I also have a 3Ghz P4 as well
even crossplatform benchmarking can't show the real difference and that because most of them are optimised and builted for one type of machines and just crossed compiled to the other, instead of rewriting the program from scratch to take care different architectures.. also different architectures can have very different results in different tests
Example: try to run distributed net OGR to that single 1.5Ghz G4 and OGR on that 3200AMD, both cores are written from scratch for the specific CPUs , results:
G4 will be about 3-4x faster than the athlon, but that is not true in everyday tasks performance
Which is why I based my statement upon running Logic Pro (called Logic Platinium at the time) on both machines... running 8 tracks on both, and then loading up the effects inserts.
I was a bit gutted at the time, as the Mac had cost me a small fortune... the Athlon64 components didn't cost more than £300 in total including 1 Gig or ram...
-
downix wrote:
bloodline wrote:
Well... err... only as far as I can see... The silverthorn, with its simple in order pipeline... but that is with good reason, i.e. to get a small die and low power consumption so it can compete with the ARM...
Not the Silverthorn, their next-gen Xeons seem to be abandoning the Core-like system and going backwards to P4-style. I think a bad move from all angles.
I already thought it'd be impossible for Intel to make good CPU's.. Now they've come to that same conclusion too :lol:
:-(
-
Speelgoedmannetje wrote:
I already thought it'd be impossible for Intel to make good CPU's.. Now they've come to that same conclusion too :lol:
:-(
There were good parts to the Pentium4 they were just outnumbered by the bad!
What they're adding back into the Xeon and Nehalem core designs are things like HyperThreading which can provide impressive performance improvements.
Andy
-
Piru wrote:
@dammy
What video cards are you using in the two machines?
What gfxcard has to do with H.264 encoding or renderings?
Any modern (current generation) graphics card can offload full HD H.264 decode from the CPU, and assist in encode.
That includes modern integrated chipsets as well. Even Intel's low-power Atom chipset has full HD decode in the chipset (licensed from Imagination, so it's PowerVR 5).
-
@Hattig
assist in encode.
Oh? Got any links to documentation?
-
No where does it say they are making a ba
Looking to the Future
Intel’s future architecture directions will continue to focus on microprocessor core enhancements, delivering performance increases and building core capabilities with improved energy efficiency. These advancements will continue to deliver compelling thresholds of energy-efficient performance that make it feasible for the same processor architecture to be used to meet the requirements of future applications across mobile, desktop, and server computing usage models.
Intel’s published roadmap provides a view of the next several generations of silicon process technology with linked architecture introductions. Intel is on track for volume production of Intel Core microarchitecture–based products on 45nm in 2007. In 2008, Intel will introduce its Nehalem (code name) microarchitecture, and the microarchitecture code-named Gesher will follow in 2010. These microarchitecture introductions will be accompanied with chipset innovations and will drive the associated platform development.
-
It depends on who you believe. Apple who lead the PPC push reversed their position in 2005 when they launched Intel models. Also some people think Altivec made a big difference.
Not sure it's true today with Core 2 Quad Extreme processors..
-
downix wrote:
Piru wrote:
@dammy
What video cards are you using in the two machines?
What gfxcard has to do with H.264 encoding or renderings?
I have an H.264 decoder in my video system here.
*edit, should specify*
It's a video decoder card I'm working on, in FPGA, plugged into one of my PCI ports, based on the newly open sourced MIT code.
Sounds cool. Any pics of your sweet project(s)?
-
AFAIK The sparc processor is something like opensource so anybody can make is own processor but i don't think it is the best processor for the amiga.
Sparc or Mips are fantastic processors but they are very specific, their future developpement are uncertain mips are unfortunately still dead, sparc is still alive but sun sell more & more x86 hardware.
I think the only mean to save Amiga is to open it to x86 to be abble to live on any PC compatible.
Amiga only lives through their early fan but who will pay thousands $ for an old/new machine ?
I really love Amiga but I don't have enough money to buy an Amiga One witch i can only use with AmigaOS 4 or Linux.
But if I can run it on a out of box PC I willl use it sure !
-
downix wrote:
Piru wrote:
If I have a 1.8GHz G5 Mac and a 2.0GHz P4, running the same OS (Say a flavour of Linux), which would yield better performance?
Depends on what you do, but G5 should beat P4 at the same clockrate at many tasks.
The thing is P4 is stoneage x86 technology these days. Take modern core2 cpu .. it runs 1333 fsb (and new ones 1600) and it runs circles around any ppc. Runs cooler, too.
But throw it against an UltraSPARC T2, and it will be crushed in both heat and performance, despite the T2 being half the clock speed.
AMD offers platforms with 780 IGP(with Radeon HD3200 GPU inc. 40 stream FP co-processors), Radeon HD34x0/36x0/38x0 and Phenom/Opteron.
Btw, Fold@Home (GPU2 client) is now available for AMD Radeon HD38x0. My AMD Radeon HD3870 has PPD rate 1746. PS3's CELL has PPD rate 900.
-
downix wrote:
Piru wrote:
If I have a 1.8GHz G5 Mac and a 2.0GHz P4, running the same OS (Say a flavour of Linux), which would yield better performance?
Depends on what you do, but G5 should beat P4 at the same clockrate at many tasks.
The thing is P4 is stoneage x86 technology these days. Take modern core2 cpu .. it runs 1333 fsb (and new ones 1600) and it runs circles around any ppc. Runs cooler, too.
But throw it against an UltraSPARC T2, and it will be crushed in both heat and performance, despite the T2 being half the clock speed.
Being out of topic, my AMD Radeon HD3870 (same as AMD FireStream 9170 but with 512MB VRAM) crushes UltraSPARC T2 in performance.
-
Piru wrote:
@dammy
What video cards are you using in the two machines?
What gfxcard has to do with H.264 encoding or renderings?
Any DX10 GPUs from AMD and NVIDIA supports H.264 1080p decoding.
http://www.hothardware.com/articles/ATI_Radeon_HD_2900_XT__R600_Has_Arrived/
AMD's Radeon HDs also supports MPEG-2, MPEG-4, DivX, WMV9, VC-1, and H.264/AVC encoding and transcoding.
-
downix wrote:
bloodline wrote:
Well... err... only as far as I can see... The silverthorn, with its simple in order pipeline... but that is with good reason, i.e. to get a small die and low power consumption so it can compete with the ARM...
Not the Silverthorn, their next-gen Xeons seem to be abandoning the Core-like system and going backwards to P4-style. I think a bad move from all angles.
Intel Nehalem is still Core 2 based.
Intel is planning for a mix of “fat” X86 and “thin” X86 processor arrays i.e. unlike CELL today, a CELL like CPU with the same ISAs.
According to AMD, DX11 specifies ray tracing.
-
downix wrote:
bloodline wrote:
Agafaster wrote:
JJ wrote:
You cant even compare different genrations of the same processor family by MHZ. MHZ is the most pointless indicator of chip performace of chips thers is. Unless you are comparing chips of eexactly the same architecture. Obivously 50mhz 030 is faster than a 25mhz 030. But you cant compare it against anything else by the MHZ
Although saying that, I do recall a 'rule of thumb' that a PPC of a certain clock could perform around 2x the equivalently clocked contemporary Pentium. I guess that'd be the PIII cf. the G3 though...
No, that would be against the Pentium4 which had very long (22 stage?) pipelines, so that very high clock speeds could be achieved, at the expense of work that could be done per clock cycle. It was a strategy based on the idea that transistor switching speeds would increase dramatically in a short space of time... this did not happen and they have currently topped out at around ~3Ghz.
AMD and Motorola opted for shorter pipelines, which resulted in lower clock speeds but more work gets done per cycle... With the Core2 architecture Intel have adopted the same approach, and brought with them all the good stuff from the P4 (ie great branch predictors and macroop fusion, etc)... (and borrowed all the good ideas of the Athlon64 and the PIII too)...
With the irony being that Intel's next-gen chips are going back to the classic P4 methodology...
Not quite i.e. shorter pipeline and a proper 128bit SIMD units.
-
Piru wrote:
@Hattig
assist in encode.
Oh? Got any links to documentation?
http://ati.amd.com/products/radeonhd2900/specs.html
"MPEG-2, MPEG-4, DivX, WMV9, VC-1, and H.264/AVC encoding and transcoding" - AMD
The new Fold@Home GPU2 client runs on the AMD’s new CAL for R6x0 GPUs.
http://www.crhc.uiuc.edu/IMPACT/ftp/talks/toronto-11-29-2007.pdf
Some NVIDIA CUDA applications i.e. H.264,RC5-72
-
Ok, the "any modern (current generation) graphics card" referred to decoding only (which I wasn't questioning).
It's still great to see that encoding is getting HW assisted, now, too. Finally some good use for those expensive 3d cards.. ;-)
[EDIT] Hmm tried to google a bit (but failed), got any links to some nice video encoder software that ulitizes the 8800GT for encoding? [/EDIT]
-
CUDA <3 (http://en.wikipedia.org/wiki/CUDA)
-
Not quite i.e. shorter pipeline and a proper 128bit SIMD units.
Yeah, I doubt they'll go back to the 31-stage pipeline of the Prescotts.
-
Piru wrote:
Ok, the "any modern (current generation) graphics card" referred to decoding only (which I wasn't questioning).
It's still great to see that encoding is getting HW assisted, now, too. Finally some good use for those expensive 3d cards.. ;-)
[EDIT] Hmm tried to google a bit (but failed), got any links to some nice video encoder software that ulitizes the 8800GT for encoding? [/EDIT]
http://www.elementaltechnologies.com/products.php?id=5
http://www.elementaltechnologies.com/news.php?id=9
To quote;
"Elemental Technologies Inc. (ETI), a leading provider of massively parallel video processing software, announced today that its new GPU-accelerated RapiHD™ H.264/AVC Encoder Plug-in for Adobe® Premiere® Pro offers up to 700 percent better performance than conventional CPU-only solutions. Elemental Technologies will demonstrate the RapiHD™ H.264/AVC Encoder Plug–in at the NAB show in Las Vegas this week"