R/E/P Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: 1 2 3 [4]   Go Down

Author Topic: AMD or Intel  (Read 23693 times)

Phillip Graham

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 280
Re: AMD or Intel
« Reply #45 on: May 29, 2006, 09:09:08 PM »

danlavry wrote on Fri, 28 April 2006 19:37


I do not have much of a preference, perhaps because I am not too deeply into understanding of PC processing. I did find myself very surprised in one area regarding processing:

I looked around to see what is available now. I expected to see some significant clock speed increase. I was hoping for some 3-5GHz clock speed, at least some X2 improvement. What happened to Moor's law? I guess it broke. I see many laptop machines with clock speeds around 2GHz, and not much higher.

Yes I know, there is dual-core architecture, and that is great for multitasking and more. But I need a fast single task machine for tasks such as compiling code, math analysis... where multitasking does not help a bit.

Am I missing something?

Regards
Dan Lavry


Hey Dan, first let me say that I only read the original thread as far as your post, so my apologies if this has nothing to do with the later discussion...

In college I interned for AMD in their Austin FAB (now a part of Spansion).  At the time (2001) we were making the original Athlons and flash RAM.  I worked in photolithography, so I got to see a pretty complete picture of the whole process (for those that don't know, photolithography is basically every-other step in making a chip).  

I was fully qualified in their online C44/50 process flow, and did several tours for VP-types of the fab, as I knew the process flow better than the other fab interns, so I have a decent background on the whole fighting moore's law thing.

I too own a IBM laptop, with a Intel chip.  This is not a "go AMD" post.

There are several key limiting steps from a physics perspective in terms of chip performance.  You have limits in conductor mobility at the transistor gate, transistor well density limits, gate oxide leakage current limits, parasistic capacitance issue in the metal layers.

To increase carrier mobility, you either have to shorten the gate  channel width, or use a different materials system (e.g. "strained" silicon germanium substrates).

To increase the capacitive coupling of the gate oxide, you need it to be as thin as possible, but then you have large leakage currents due to tunneling.  The gate oxide needs to be epitaxial though, to minimize charge traps at the depletion channel interface.  This was actually the original reason for picking silicon as the substrate.  It could be grown single crystal, and had a stable epitaxial gate oxide.

Finally, you have to embed each transitor in its own little oxide "box" to prevent latchup between adjacent transistors.  Gate leakage and transistor to transistor leakage are both big power drains.  Silicon on insulator (SOI) helps a lot with the the latter, the former is trickier.  IBM, AMD, and (possibly?) Intel are now using SOI substrates in at least some capacity.

Then you have the problem of practical fabrication below 193nm lithographic techniques.  Excimer sources get harder to make, spinning the photoresists thin enough to develop effectively w/ the super short wavelength lithography is extremely hard, air adsorption becomes appreciable, etc.

Then you have the problem of the electricity's speed in the metallization layers above the transistors.  The capacitance of the insulating layers slows the efield propagation.  This was the achilles heel of the old athlon.

Sorry, must leave this post unfinished, gotta run, will try to finish it later...
Logged
Phillip Graham

kraster

  • Full Member
  • ***
  • Offline Offline
  • Posts: 199
Re: AMD or Intel
« Reply #46 on: May 30, 2006, 06:21:51 AM »

Phillip Graham wrote on Tue, 30 May 2006 02:09

danlavry wrote on Fri, 28 April 2006 19:37


I do not have much of a preference, perhaps because I am not too deeply into understanding of PC processing. I did find myself very surprised in one area regarding processing:

I looked around to see what is available now. I expected to see some significant clock speed increase. I was hoping for some 3-5GHz clock speed, at least some X2 improvement. What happened to Moor's law? I guess it broke. I see many laptop machines with clock speeds around 2GHz, and not much higher.

Yes I know, there is dual-core architecture, and that is great for multitasking and more. But I need a fast single task machine for tasks such as compiling code, math analysis... where multitasking does not help a bit.

Am I missing something?

Regards
Dan Lavry


Hey Dan,

In college I interned for AMD in their Austin FAB (now a part of Spansion).  At the time (2001) we were making the original Athlons and flash RAM.  I worked in photolithography, so I got to see a pretty complete picture of the whole process (for those that don't know, photolithography is basically every-other step in making a chip).  

I was fully qualified in their online C44/50 process flow, and did several tours for VP-types of the fab, as I knew the process flow better than the other fab interns, so I have a decent background on the whole fighting moore's law thing.

I too own a IBM laptop, with a Intel chip.  This is not a "go AMD" post.

There are several key limiting steps from a physics perspective in terms of chip performance.  You have limits in conductor mobility at the transistor gate, transistor well density limits, gate oxide leakage current limits, parasistic capacitance issue in the metal layers.

To increase carrier mobility, you either have to shorten the gate  channel width, or use a different materials system (e.g. "strained" silicon germanium substrates).

To increase the capacitive coupling of the gate oxide, you need it to be as thin as possible, but then you have large leakage currents due to tunneling.  The gate oxide needs to be epitaxial though, to minimize charge traps at the depletion channel interface.  This was actually the original reason for picking silicon as the substrate.  It could be grown single crystal, and had a stable epitaxial gate oxide.

Finally, you have to embed each transitor in its own little oxide "box" to prevent latchup between adjacent transistors.  Gate leakage and transistor to transistor leakage are both big power drains.  Silicon on insulator (SOI) helps a lot with the the latter, the former is trickier.  IBM, AMD, and (possibly?) Intel are now using SOI substrates in at least some capacity.

Then you have the problem of practical fabrication below 193nm lithographic techniques.  Excimer sources get harder to make, spinning the photoresists thin enough to develop effectively w/ the super short wavelength lithography is extremely hard, air adsorption becomes appreciable, etc.

Then you have the problem of the electricity's speed in the metallization layers above the transistors.  The capacitance of the insulating layers slows the efield propagation.  This was the achilles heel of the old athlon.

Sorry, must leave this post unfinished, gotta run, will try to finish it later...


That's funny. I was just thinking the same thing!!!! Shocked  Very Happy

All I can say is wow.
Logged

Phillip Graham

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 280
Re: AMD or Intel
« Reply #47 on: May 31, 2006, 09:36:49 PM »

Phillip Graham wrote on Mon, 29 May 2006 21:09

danlavry wrote on Fri, 28 April 2006 19:37


I do not have much of a preference, perhaps because I am not too deeply into understanding of PC processing. I did find myself very surprised in one area regarding processing:

I looked around to see what is available now. I expected to see some significant clock speed increase. I was hoping for some 3-5GHz clock speed, at least some X2 improvement. What happened to Moor's law? I guess it broke. I see many laptop machines with clock speeds around 2GHz, and not much higher.

Yes I know, there is dual-core architecture, and that is great for multitasking and more. But I need a fast single task machine for tasks such as compiling code, math analysis... where multitasking does not help a bit.

Am I missing something?

Regards
Dan Lavry


Hey Dan, first let me say that I only read the original thread as far as your post, so my apologies if this has nothing to do with the later discussion...

In college I interned for AMD in their Austin FAB (now a part of Spansion).  At the time (2001) we were making the original Athlons and flash RAM.  I worked in photolithography, so I got to see a pretty complete picture of the whole process (for those that don't know, photolithography is basically every-other step in making a chip).  

I was fully qualified in their online C44/50 process flow, and did several tours for VP-types of the fab, as I knew the process flow better than the other fab interns, so I have a decent background on the whole fighting moore's law thing.

I too own a IBM laptop, with a Intel chip.  This is not a "go AMD" post.

There are several key limiting steps from a physics perspective in terms of chip performance.  You have limits in conductor mobility at the transistor gate, transistor well density limits, gate oxide leakage current limits, parasistic capacitance issue in the metal layers.

To increase carrier mobility, you either have to shorten the gate  channel width, or use a different materials system (e.g. "strained" silicon germanium substrates).

To increase the capacitive coupling of the gate oxide, you need it to be as thin as possible, but then you have large leakage currents due to tunneling.  The gate oxide needs to be epitaxial though, to minimize charge traps at the depletion channel interface.  This was actually the original reason for picking silicon as the substrate.  It could be grown single crystal, and had a stable epitaxial gate oxide.

Finally, you have to embed each transitor in its own little oxide "box" to prevent latchup between adjacent transistors.  Gate leakage and transistor to transistor leakage are both big power drains.  Silicon on insulator (SOI) helps a lot with the the latter, the former is trickier.  IBM, AMD, and (possibly?) Intel are now using SOI substrates in at least some capacity.

Then you have the problem of practical fabrication below 193nm lithographic techniques.  Excimer sources get harder to make, spinning the photoresists thin enough to develop effectively w/ the super short wavelength lithography is extremely hard, air adsorption becomes appreciable, etc.

Then you have the problem of the electricity's speed in the metallization layers above the transistors.  The capacitance of the insulating layers slows the efield propagation.  This was the achilles heel of the old athlon.

Sorry, must leave this post unfinished, gotta run, will try to finish it later...


It appears the continuation must be in a new post for the BB to bounce the continued reply to the top.  I have included the initial response as a quote for clarity.

On with show...

So, as the previous post exposited, there are numerous factors in play here for the speed of the chips.

My internship with AMD ended a few weeks before 9/11, and before the planned "tape out" of the initial Athlon64 (Sledgehammer and Clawhammer).  The K8/Athlon64/xxxHammer is a totally different beast, and I don't know any details about it.

Intel was totally kicking AMDs butt megahertz-wise at that time, even with some proprietary tricks AMD used to make smaller transistors.  They had slow spots in the metal layers that lead down to the logic block level.  Improving the speed of those chips came mostly from reworking the "back end" of the process, as the metal layers are called.

Fast-forward to today, and the latest chips.  The smaller you make your transitors (the shorter the depletion channel), the faster you can pump them on and off.  The better you capacitively couple the gate to the transitor, the better they respond on/off at lower voltages.  The smaller you make each transitor, the more you can pack together.

Of course all those close together transistors all have "leaking" source/drain currents, gate currents, and all other mess.  Pretty soon you have a BUNCH of current flowing through your chip that is not directly related to "on/off" states.  If your chip has a core voltage of 2V, and is disspating 50+ watts of heat, it doesn't take a rocket scientist to see that a ton of current is flowing through the chip.

It is no longer simply a function of "make everything smaller" which has it's own set of new challenges coming into view (e.g. vacuum lithography), but also the lackluster properties of silicon/silicon dioxide, especially in heat production.

Silicon is not a direct bandgap semiconductor; the path from valence to conduction band, in reciprocal space, has a tangential component that causes phonon formation (laymans--electrons moving in silicon create packets of lattice vibrations, i.e. heat).  Silicon is also not a wide bandgap semiconductor, but that is less of a concern for microprocessors.  Silicon dioxide is a low K dielectric, and has a fair loss tangent.  It also has less than awesome dielectric breakdown behavior.

So, you can scrap 40 years of chemistry and solid state physics invested in silicon, and try to find a replacement materials system (and many are working on this), or you can see that end of practical chips made with higher transitor density in plain silicon is drawing to a close, and start making multiple cores on the same bit of wafer.  That is what is happening now.

Of course that is not a panacea either, as multiple cores makes for bigger chips, unless you also increase the transitor density.  So, in the end there is no free lunch, and silicon, as a semiconductor system, is pretty long in the tooth.

I think many of these problems have really come to a head for the manufacturers in the past couple years.  AMD has been able to catch up somewhat in terms of fabrication ability in recent times, and intel is hampered, currently, but several less than stellar chip architectures.

Please feel free to ask any questions these messages have spurred.

As a bit of spin to an otherwise highly technical post, I would put my .02 in here and say that i feel AMD, at THIS time, is making a superior product to Intel.  I don't have any AMD or intel stock, and several of my close friends work for intel, so I am not saying that out of blind enthusiasm for my past employer.  AMD certainly has not always been in this position, and I am glad to see their success, as well as competition in the industry, which only benefits the consumers.

As a final useful bit for all you DIY PC people out there, AMDs internal reference motherboards for testing (in 2k1) were built by Tyan, because they had the best motherboard design/quality by AMDs judgement.
Logged
Phillip Graham

kraster

  • Full Member
  • ***
  • Offline Offline
  • Posts: 199
Re: AMD or Intel
« Reply #48 on: May 31, 2006, 10:42:55 PM »

Excellent post again. Can you elaborate on the alternative materials that are being developed that you mentioned in your post?

Fascinating stuff (the little I understand.) Wink
Logged

Phillip Graham

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 280
Re: AMD or Intel
« Reply #49 on: June 01, 2006, 01:45:18 AM »

kraster wrote on Wed, 31 May 2006 22:42

Excellent post again. Can you elaborate on the alternative materials that are being developed that you mentioned in your post?

Fascinating stuff (the little I understand.) Wink



Well, first let me apologize if the posts are too technical.  I was responding to Mr. Lavry, and presumed, perhaps incorrectly, that he has formal training as an electrical engineer.  EEs typically get at least one semester of solid state physics in college, so I figured that some of the post would ring a bell.  Of course, who knows, I have already forgotten tons of my college education....

As for alternative systems, there are quite a few "wide band gap" systems under investigation.  GaAs and SiC (Gallium Arsenide and Silicon Carbide) are the first two that come to mind for some reason.

I could talk a little more about why they are better, but it requires at least understanding what a "band gap" is.  So I start by asking if you have heard the term and/or know what it means.

Logged
Phillip Graham

crm0922

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 272
Re: AMD or Intel
« Reply #50 on: June 01, 2006, 08:29:11 AM »

Tyan eh? Wink

Is SiGe and other strained silicon approaches viable for CPU semiconductor material in the future or is that just another stopgap?

I am actually of friend of the so-called "father" of strained silicon high mobility science.  I should point him to this thread, that would be funny (and probably informative if he posted a response).

Chris

Logged

Phillip Graham

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 280
Re: AMD or Intel
« Reply #51 on: June 01, 2006, 08:55:06 AM »

crm0922 wrote on Thu, 01 June 2006 08:29

Tyan eh? Wink

Is SiGe and other strained silicon approaches viable for CPU semiconductor material in the future or is that just another stopgap?

I am actually of friend of the so-called "father" of strained silicon high mobility science.  I should point him to this thread, that would be funny (and probably informative if he posted a response).

Chris




Hey Chris,

I don't know if SiGe ('siggee') has made any real inroads on microprocessors, it does do a pretty nice job of increasing the carrier mobility.  I took an AMD internal course on it during my internship, but that was the extent of my exposure.

I know an RF amplifier guy who designs in SiGe, but that is for a bleeding edge fabless startup type company, so your guess is as good as mine in the big picture.
Logged
Phillip Graham

danlavry

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 997
Re: AMD or Intel
« Reply #52 on: June 01, 2006, 03:10:29 PM »

Phillip Graham wrote on Thu, 01 June 2006 06:45

kraster wrote on Wed, 31 May 2006 22:42

Excellent post again. Can you elaborate on the alternative materials that are being developed that you mentioned in your post?

Fascinating stuff (the little I understand.) Wink



Well, first let me apologize if the posts are too technical.  I was responding to Mr. Lavry, and presumed, perhaps incorrectly, that he has formal training as an electrical engineer.  EEs typically get at least one semester of solid state physics in college, so I figured that some of the post would ring a bell.  Of course, who knows, I have already forgotten tons of my college education....

As for alternative systems, there are quite a few "wide band gap" systems under investigation.  GaAs and SiC (Gallium Arsenide and Silicon Carbide) are the first two that come to mind for some reason.

I could talk a little more about why they are better, but it requires at least understanding what a "band gap" is.  So I start by asking if you have heard the term and/or know what it means.




Hi Phillip,

It is just fine to have your comments here. Some can handle it, others can not.

Yes, I have a formal training, and yes I recall very well the fundamentals of semiconductors. I still have my old books such as "Solid State Electronics" by Wang, dated 1966... I probably learned more for 3 books by Motorola (Motorola series in solid state electronics). One was "Analysis and Design of integrated circuits" 1968, but the one that would apply more to the subject here is "Integrated circuits, Design Principles and fabrications" 1965. Needless to say, all that stuff is old. Some of the principles are the same. When I started real design work, CMOS was the "new thing" on the block.  

My comment here is: After graduation, I quickly realized that I have to "draw a line in the sand". An EE can dedicate his career to be focused on one aspect of electronics, or another aspect... I had to make choices: focus on semiconductor manufacturing (including the physics and chemistry of it), focus on circuits and equipment design, or focus on something else. I was there early enough to be able to do both analog and digital circuit design. That took a full time commitment, and I did have to "let go" of any deep pursuit of what goes into making transistors and IC's. In fact, I "specialized" in "understanding the end result" from a "user standpoint" - reading and understanding the data sheets in terms of voltage, current, capacitance, heat....

Today, a new graduate has to limit their scope a lot tighter! Some focus on computer hardware, others on software. A few get in to analog design, a few into semiconductor processing....
I come from the tube era, so I grew with much of the electronics. The new EE's are forced to much tighter specialization...

Yet, it should come to you as no surprise that an EE does not get to know all aspects of electronics. I am generally aware of some factors that have to do with speed, but certainly to a much lesser degree then an EE in a semiconductor manufacturing facility. My question why the processor speed seem to have "slowed down" lately is an OK question, in context of my area's of expertize, and the area's that I did not choose to pursue.

Your technical comments are very welcome. Of course I can not be much of a moderator about what makes a processor work faster. Remember, I was the one that started this thread, and it begun with a question.

Regards
Dan Lavry
http://www.lavryengineering.com  
Logged

Phillip Graham

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 280
Re: AMD or Intel
« Reply #53 on: June 05, 2006, 11:45:04 AM »

danlavry wrote on Thu, 01 June 2006 15:10


Hi Phillip,

It is just fine to have your comments here. Some can handle it, others can not.

Today, a new graduate has to limit their scope a lot tighter! Some focus on computer hardware, others on software. A few get in to analog design, a few into semiconductor processing....
I come from the tube era, so I grew with much of the electronics. The new EE's are forced to much tighter specialization...


Yeah, tight specialization seems to be the norm, like my friend who designs RF amplifiers needing to do a lot of moonlight reading to learn those, even with his masters.  I've also met guys who were great with DSP, but could not understand very well the passive XO in a speaker I would be putting together.

I wish I knew more about the design side, but you can't learn everything.  Seeing how you go from an inverter to boolean was enough of a lightbulb for me!

Quote:


Your technical comments are very welcome. Of course I can not be much of a moderator about what makes a processor work faster. Remember, I was the one that started this thread, and it begun with a question.

Regards
Dan Lavry
http://www.lavryengineering.com  


Hey Dan, I wrote my reply with you specifically in mind and figured a fairly high level of reply was appropriate.  Many different things have sort of balled together to make moores law more difficult in the past couple of years.  I am sure some of it comes from the layout side, but much of it is device physics issues.
Logged
Phillip Graham

danlavry

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 997
Re: AMD or Intel
« Reply #54 on: August 28, 2006, 07:50:59 PM »

crm0922 wrote on Thu, 01 June 2006 13:29

Tyan eh? Wink

Is SiGe and other strained silicon approaches viable for CPU semiconductor material in the future or is that just another stopgap?

I am actually of friend of the so-called "father" of strained silicon high mobility science.  I should point him to this thread, that would be funny (and probably informative if he posted a response).

Chris




Well, here are some real wold test results:

Computer A: My older laptop, an IBM thinkpad laptop, model A51, about 1.8MHz clock speed, running under Win2000.

Computer B: My newer laptop, an IBM thinkpad (Lenovo)laptop, model X60, about 1.83MHz clock speed, running under Win2000.

Both have similar amount of memory (about 500Meg) and other similar details (the X60 has somewhat faster front bus speed, but not by much if I recall...). The A51 is running a pentium 4. The X60 is running a Intel centrino duo.

For the test, I was running a compiler for an Altera FPGA, and then I ran a "timing diagram" simulation test.

Results:
Computer A: Compile time 39 seconds, Simulation time 8 seconds
Computer B: Compile time 17 seconds, Simulation time 5 seconds
 
That was of course one small circuit, but it is representative of what I do. So I am sold on the fact that processor speed is not telling much of the story. The X60 is about twice as fast in the real world, at least in my real world (where video speed does not matter much)...

Regards
Dan Lavry
http://www.lavryengineering.com
Logged

crm0922

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 272
Re: AMD or Intel
« Reply #55 on: August 29, 2006, 02:34:13 PM »

I think the dual core is giving you some help here, it looks like around 1.6X improvement.

What software is it specifically?  The drive speed on laptops is far slower than desktops in most configurations.  And don't forget that the Centrino platform is not designed for speed as much as it is for portability and low power consumption.

Intel's site says:


Enjoy greater than 2X CPU performance and up to 28 percent power reduction with the new Intel
Logged

Andy Peters

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1124
Re: AMD or Intel
« Reply #56 on: August 29, 2006, 03:02:23 PM »

danlavry wrote on Mon, 28 August 2006 16:50

Well, here are some real wold test results:

Computer A: My older laptop, an IBM thinkpad laptop, model A51, about 1.8MHz clock speed, running under Win2000.

Computer B: My newer laptop, an IBM thinkpad (Lenovo)laptop, model X60, about 1.83MHz clock speed, running under Win2000.

Both have similar amount of memory (about 500Meg) and other similar details (the X60 has somewhat faster front bus speed, but not by much if I recall...). The A51 is running a pentium 4. The X60 is running a Intel centrino duo.

For the test, I was running a compiler for an Altera FPGA, and then I ran a "timing diagram" simulation test.

Results:
Computer A: Compile time 39 seconds, Simulation time 8 seconds
Computer B: Compile time 17 seconds, Simulation time 5 seconds
 
That was of course one small circuit, but it is representative of what I do. So I am sold on the fact that processor speed is not telling much of the story. The X60 is about twice as fast in the real world, at least in my real world (where video speed does not matter much)...


Windows-based FPGA tools unfortunately are not designed to take advantage of dual processors.  However, what I've noticed with the Xilinx tools on this hyperthreaded P4 Dell POF here is that one of the "CPUs" will run the FPGA tools and the other tends to handle the user interface.  I'd expect a true dual-core machine to act the same way (and I'll test it once I get Boot Camp running on my Mac Book Pro).

BTW: 512 MB ain't enough for FPGA tools.  Max that machine out.  You want as much of the design in real memory as possible.  Also, I've brought machines to their knees with even moderate-sized FPGA simulations (using ModelSim SE), but then again my test benches are fairly complex.  (No schematic, all HDL.)

-a
Logged
"On the Internet, nobody can hear you mix a band."

danlavry

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 997
Re: AMD or Intel
« Reply #57 on: August 30, 2006, 02:14:23 PM »

crm0922 wrote on Tue, 29 August 2006 19:34

I think the dual core is giving you some help here, it looks like around 1.6X improvement.

What software is it specifically?  The drive speed on laptops is far slower than desktops in most configurations.  And don't forget that the Centrino platform is not designed for speed as much as it is for portability and low power consumption.

Intel's site says:


Enjoy greater than 2X CPU performance and up to 28 percent power reduction with the new Intel? Centrino? Duo mobile technology laptops based on Intel? Coreā„¢ Duo processor as compared to previous generation Intel? Centrino? mobile technology-based laptops.+



And that is basically what we've seen here, minus the marketing hype (greater than 2X...not quite).

Chris


Hi,

The test was done on the Quartus software by Altera. It was not a huge circuit compile/simulate test, but enough for me to be convinced with that small circuit example.

Yes, the power consumption and the weight of the X60 were a major consideration for me. The weight is very small because the machine "comes apart" from the "base" which contains the CD rom and various ports..

Regards
Dan Lavry
http://www.lavryengineering.com    
Logged

danlavry

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 997
Re: AMD or Intel
« Reply #58 on: August 30, 2006, 02:44:45 PM »

Andy Peters wrote on Tue, 29 August 2006 20:02

danlavry wrote on Mon, 28 August 2006 16:50

Well, here are some real wold test results:

Computer A: My older laptop, an IBM thinkpad laptop, model A51, about 1.8MHz clock speed, running under Win2000.

Computer B: My newer laptop, an IBM thinkpad (Lenovo)laptop, model X60, about 1.83MHz clock speed, running under Win2000.

Both have similar amount of memory (about 500Meg) and other similar details (the X60 has somewhat faster front bus speed, but not by much if I recall...). The A51 is running a pentium 4. The X60 is running a Intel centrino duo.

For the test, I was running a compiler for an Altera FPGA, and then I ran a "timing diagram" simulation test.

Results:
Computer A: Compile time 39 seconds, Simulation time 8 seconds
Computer B: Compile time 17 seconds, Simulation time 5 seconds
 
That was of course one small circuit, but it is representative of what I do. So I am sold on the fact that processor speed is not telling much of the story. The X60 is about twice as fast in the real world, at least in my real world (where video speed does not matter much)...


Windows-based FPGA tools unfortunately are not designed to take advantage of dual processors.  However, what I've noticed with the Xilinx tools on this hyperthreaded P4 Dell POF here is that one of the "CPUs" will run the FPGA tools and the other tends to handle the user interface.  I'd expect a true dual-core machine to act the same way (and I'll test it once I get Boot Camp running on my Mac Book Pro).

BTW: 512 MB ain't enough for FPGA tools.  Max that machine out.  You want as much of the design in real memory as possible.  Also, I've brought machines to their knees with even moderate-sized FPGA simulations (using ModelSim SE), but then again my test benches are fairly complex.  (No schematic, all HDL.)

-a


The laptop is for "work on the go" such as at home in the evening, and my desktop is just too heavy to carry Smile
My FPGA and CPLD work is a combination of VHDL and schematic entry, but I tend to break the tasks into smaller chunks, and that certainly is helpfull. Of course, at some point one needs to put it all together.

I agree that the software can "bring a machine down to it's knees". That is one reason why I do what I can to break it to smaller jobs. There are almost always some changes to be made, and I just hate it to have to compile a whole big job because of some little thing during development. That method (breaking to smaller tasks) does not always work out (such as in cases that are speed or layout critical), but it works most of the time, mostly for checking concepts and building blocks).

But I do many other things on the laptops, such as schematics, printed circuit layout, math and much of the mechanical design, for all my products.

The difference in weight between the A51 and X60 is huge, and the difference in battery useage time (before recharge) is also huge. That is why I bought the X60. The speed is a nice surprise for me.

The Altera FPGA software will be faster soon, when it gets converted to 64 bit capability.

Regards
Dan Lavry
http://www.lavryengineering.com

 

Logged
Pages: 1 2 3 [4]   Go Up