2008/9/25 Chris Kassopulo <[hidden email]>:
> > http://www.lesser-software.com/en/content/products/lswvst/lswvst.htm > > It would be nice to look at it (LSWV). So many cool features. Too bad its proprietary. :( -- Best regards, Igor Stasenko AKA sig. |
Which brings me right back to Simon Kirk's post a few days ago...! > >http://www.robvens.nl/en/blog/The-Smalltalk-Trap.html > >Cheers, >Simon I'm not saying everything should be free, though. It's just an interesting problem... Rob On Thu, Sep 25, 2008 at 3:15 PM, Igor Stasenko <[hidden email]> wrote: 2008/9/25 Chris Kassopulo <[hidden email]>: |
In reply to this post by Igor Stasenko
On Thu, 25 Sep 2008 21:15:18 +0200, Igor Stasenko wrote:
> 2008/9/25 Chris Kassopulo : >> >> http://www.lesser-software.com/en/content/products/lswvst/lswvst.htm >> > > It would be nice to look at it (LSWV). > So many cool features. Too bad its proprietary. :( What would be the propriety portion of Frank's Smalltalk and its combination with C#, C++, C, Prolog and Java for .NET MS-Windows, Unix-Systems and embedded systems? I talked with Frank at some Smalltalk happening recently, he conducts a business model with paying customers, who could be very "nervous" if suddenly his Smalltalk would be available for free. In comparision, Squeak has quite the opposite business model (no offense). You managers (and wanna-be managers) step forward, any ideas for how the larger part of Smalltalk community could benefit? /Klaus |
Hi!,
Diversity of bussiness models is granted by nature. All bussiness models are granted to coexist. The reduction of medium/small bussiness space in the last decade (canivalization made by big companies and free-workers reducing the small-bussiness space) can revert in the next years (I hope :-) Now, we can see some actors reacting to current idealized/abstracted/reduced world. To make the products of our work free is not the best way to hide sub-emploiment. (no offense) > You managers (and wanna-be managers) step forward, > any ideas for how the > larger part of Smalltalk community could benefit? Diverse options to make bussiness is important to managers to select the best model for each production context. IMO, Smalltalk community do not need to be promoted, it is a concecuence or our shared activities (it is not an object :). cheers, Ale. |
In reply to this post by Igor Stasenko
On Thu, Sep 25, 2008 at 9:15 PM, Igor Stasenko <[hidden email]> wrote:
> 2008/9/25 Chris Kassopulo <[hidden email]>: >> >> http://www.lesser-software.com/en/content/products/lswvst/lswvst.htm >> >> > > It would be nice to look at it (LSWV). > So many cool features. Too bad its proprietary. :( > > -- > Best regards, > Igor Stasenko AKA sig. Is the cost prohibitive? Dolphin wasn't free, but it was really cheap and a quality product. Keep in mind that while programming languages have been largely "commoditized" the best implementations of many languages still cost money. E.g. you want the best C compiler? It's not GCC. Buy the Intel compiler and watch your code run twice as fast. |
Jason Johnson escreveu:
Big shortcoming of Dolphin is that it is not cross platform (I mean, it only runs in Micro$oft Window$). I do prefer Cincom suite.On Thu, Sep 25, 2008 at 9:15 PM, Igor Stasenko [hidden email] wrote:2008/9/25 Chris Kassopulo [hidden email]:http://www.lesser-software.com/en/content/products/lswvst/lswvst.htmIt would be nice to look at it (LSWV). So many cool features. Too bad its proprietary. :( -- Best regards, Igor Stasenko AKA sig.Is the cost prohibitive? Dolphin wasn't free, but it was really cheap and a quality product. Keep in mind that while programming languages have been largely "commoditized" the best implementations of many languages still cost money. E.g. you want the best C compiler? It's not GCC. Buy the Intel compiler and watch your code run twice as fast. But I felt like this list is for squeak development discussions... And there are lots of issues in this field (squeak development). Starting to discuss how to make its VM preemptive, then real multitasking, then really suited for distributed processing... -- ############################################################################### # Este e-mail pode conter informações confidenciais/privadas e # destina-se somente aos destinatários especificados no cabeçalho # (campos To:, Cc:, CCo:) # # O repasse de parte ou da totalidade deste e-mail para outros # usuários ou para listas de discussão deve ser autorizado # explicitamente por [hidden email] # # Esta mensagem é digitalmente assinada utilizando-se algorítimo # PGP (gnuPG). A chave pública para o usuário [hidden email] # pode ser obtida em: http://pgp.mit.edu # # ----------------------------------------------------------------------------- # # This message may contain confidential/private information and # is directed only to the recipients specified in the message # header (fields To:, Cc:, CCo:). # # Forwarding part or the totality of this message to other people # requires explicit authorization from [hidden email] # # This message is digitally signed using GnuPG (PGP algorithm). Public # key for [hidden email] may be recovered from: # http://pgp.mit.edu # ############################################################################### signature.asc (267 bytes) Download Attachment |
In reply to this post by Jason Johnson-5
2008/9/26, Jason Johnson <[hidden email]>:
> On Thu, Sep 25, 2008 at 9:15 PM, Igor Stasenko <[hidden email]> wrote: >> 2008/9/25 Chris Kassopulo <[hidden email]>: >>> >>> http://www.lesser-software.com/en/content/products/lswvst/lswvst.htm >>> >>> >> >> It would be nice to look at it (LSWV). >> So many cool features. Too bad its proprietary. :( >> >> -- >> Best regards, >> Igor Stasenko AKA sig. > > Is the cost prohibitive? Dolphin wasn't free, but it was really cheap > and a quality product. Keep in mind that while programming languages > have been largely "commoditized" the best implementations of many > languages still cost money. E.g. you want the best C compiler? It's > not GCC. Buy the Intel compiler and watch your code run twice as > fast. If you define best best as "longest bar in some benchmark". If you want to run your software only on Intel chips, no AMD, no Motorola, no Sun, no .... If you want to put up with all kinds of trouble and issues compiling software that was written with GCC in mind like all the GNU/Linux software. Maybe even Squeak thanks to gnuify. Yeah, then ICC might be the "best". Cheers Philippe |
In reply to this post by Jason Johnson-5
> Keep in mind that while programming languages
> have been largely "commoditized" the best implementations of many > languages still cost money. E.g. you want the best C compiler? It's > not GCC. Buy the Intel compiler and watch your code run twice as > fast. wrong, according to the people there (from google "icc vs gcc", first results page): http://blog.alphagemini.org/2008/03/icc-vs-gcc-43.html http://www.osnews.com/comments/19462 business is business, quality is quality, saying business => quality is an ideological statement that needs yet to be proven. IMHO :) Stef |
In reply to this post by Philippe Marschall
2008/9/26, Jason Johnson [hidden email]:On Thu, Sep 25, 2008 at 9:15 PM, Igor Stasenko [hidden email] wrote:2008/9/25 Chris Kassopulo [hidden email]:http://www.lesser-software.com/en/content/products/lswvst/lswvst.htmIt would be nice to look at it (LSWV). So many cool features. Too bad its proprietary. :( -- Best regards, Igor Stasenko AKA sig.Is the cost prohibitive? Dolphin wasn't free, but it was really cheap and a quality product. Keep in mind that while programming languages have been largely "commoditized" the best implementations of many languages still cost money. E.g. you want the best C compiler? It's not GCC. Buy the Intel compiler and watch your code run twice as fast. Let me preface this by saying that I support free software, and use it in preference to proprietary software when possible. However, I disagree with much of what you say. If you define best best as "longest bar in some benchmark". I define best as "producing the best performance for the program that I am interested in compiling". Intel doesn't manage to charge serious money for their compiler because it is better on some synthetic benchmark but worse in the real world. Maybe GCC has mostly caught up with 4.3, but there's no doubt that ICC has traditionally generated faster code. I don't have extensive/varied personal experience with ICC, but if you compile Squeak with it on Intel, you'll see a 20% speed-up on macro benchmarks. If you want to run your software only on Intel chips, no AMD, no Motorola, no Sun, no .... First, it's simply not true that Intel's compiler doesn't work for AMD. After surfing around for 15min or so, I learned: - at various times in the past, AMD chose ICC as the compiler they used to generate SPEC benchmarks. - there are numerous personal accounts where ICC generates the fastest code for AMD for their pet application. Typically, Intel CPUs have a higher percentage improvement by using ICC instead of GCC, but AMD CPUs also benefit. Also, many applications don't target PowerPC, Sparc, or ARM. For desktop apps, x86 is the name of the game. I don't feel limited by using an x86-only compiler, especially since I can write my code to compile with a different compiler for other platforms. If you want to put up with all kinds of trouble and issues compiling software that was written with GCC in mind like all the GNU/Linux software. That's an interesting angle. Many Free software lovers would consider code that uses specific features of Microsoft's C compiler to be broken. I generally agree with them; unless there is a good reason for it (and there sometimes is, just as it sometimes makes sense to write x86 assembly), compiler-specific code is a bug. Why should it be any different for GNU-specific code? To me, it seems like FUD to disparage a compiler for inability to compile non-standard code. Maybe even Squeak thanks to gnuify. No, gnuify transforms the code to make it run faster on GCC, but if you don't apply gnuify, it will compile fine with other compilers. BTW, Windows Squeak VMs are still built using GCC 2.9.5 becuase it produces faster code than 4.x (don't know if 4.3 has been tried). The moral: if you care about performance, run your own benchmarks. Then, you can balance observed performance against other factors like cost, portability, free-software ideology, etc. Cheers, Josh Yeah, then ICC might be the "best". Cheers Philippe |
2008/9/27 Joshua Gargus <[hidden email]>:
> Philippe Marschall wrote: > > 2008/9/26, Jason Johnson <[hidden email]>: > > > On Thu, Sep 25, 2008 at 9:15 PM, Igor Stasenko <[hidden email]> wrote: > > > 2008/9/25 Chris Kassopulo <[hidden email]>: > > > http://www.lesser-software.com/en/content/products/lswvst/lswvst.htm > > > > > It would be nice to look at it (LSWV). > So many cool features. Too bad its proprietary. :( > > -- > Best regards, > Igor Stasenko AKA sig. > > > Is the cost prohibitive? Dolphin wasn't free, but it was really cheap > and a quality product. Keep in mind that while programming languages > have been largely "commoditized" the best implementations of many > languages still cost money. E.g. you want the best C compiler? It's > not GCC. Buy the Intel compiler and watch your code run twice as > fast. > > > > > Let me preface this by saying that I support free software, and use it in > preference to proprietary software when possible. However, I disagree with > much of what you say. > > If you define best best as "longest bar in some benchmark". > > I define best as "producing the best performance for the program that I am > interested in compiling". > > Intel doesn't manage to charge serious money for their compiler because it > is better on some synthetic benchmark but worse in the real world. Maybe > GCC has mostly caught up with 4.3, but there's no doubt that ICC has > traditionally generated faster code. I don't have extensive/varied personal > experience with ICC, but if you compile Squeak with it on Intel, you'll see > a 20% speed-up on macro benchmarks. > > If you > want to run your software only on Intel chips, no AMD, no Motorola, no > Sun, no .... > > First, it's simply not true that Intel's compiler doesn't work for AMD. > After surfing around for 15min or so, I learned: > - at various times in the past, AMD chose ICC as the compiler they used to > generate SPEC benchmarks. > - there are numerous personal accounts where ICC generates the fastest code > for AMD for their pet application. > > Typically, Intel CPUs have a higher percentage improvement by using ICC > instead of GCC, but AMD CPUs also benefit. > > Also, many applications don't target PowerPC, Sparc, or ARM. For desktop > apps, x86 is the name of the game. I don't feel limited by using an > x86-only compiler, especially since I can write my code to compile with a > different compiler for other platforms. > > If you want to put up with all kinds of trouble and > issues compiling software that was written with GCC in mind like all > the GNU/Linux software. > > That's an interesting angle. Many Free software lovers would consider code > that uses specific features of Microsoft's C compiler to be broken. I > generally agree with them; unless there is a good reason for it (and there > sometimes is, just as it sometimes makes sense to write x86 assembly), > compiler-specific code is a bug. Why should it be any different for > GNU-specific code? Such things happen automatically if you compile with only one compiler. You can observe it every time a new GCC version comes out that is more standard conformant a lot of software breaks. You can also see it with Squeak. A lot of code has empty statements and temps declared in a too wide scope. Once we get a new compiler that supports closures or you compile it with a more standard conformant compiler, that code has to be updated. > To me, it seems like FUD to disparage a compiler for inability to compile > non-standard code. > > Maybe even Squeak thanks to gnuify. > > No, gnuify transforms the code to make it run faster on GCC, but if you > don't apply gnuify, it will compile fine with other compilers. But slower. You don't actually pay for this, do you? Cheers Philippe |
In reply to this post by Joshua Gargus-2
2008/9/27 Joshua Gargus <[hidden email]>:
> Philippe Marschall wrote: > > 2008/9/26, Jason Johnson <[hidden email]>: > > > On Thu, Sep 25, 2008 at 9:15 PM, Igor Stasenko <[hidden email]> wrote: > > > 2008/9/25 Chris Kassopulo <[hidden email]>: > > > http://www.lesser-software.com/en/content/products/lswvst/lswvst.htm > > > > > It would be nice to look at it (LSWV). > So many cool features. Too bad its proprietary. :( > > -- > Best regards, > Igor Stasenko AKA sig. > > > Is the cost prohibitive? Dolphin wasn't free, but it was really cheap > and a quality product. Keep in mind that while programming languages > have been largely "commoditized" the best implementations of many > languages still cost money. E.g. you want the best C compiler? It's > not GCC. Buy the Intel compiler and watch your code run twice as > fast. > > > > > Let me preface this by saying that I support free software, and use it in > preference to proprietary software when possible. However, I disagree with > much of what you say. > > If you define best best as "longest bar in some benchmark". > > I define best as "producing the best performance for the program that I am > interested in compiling". > > Intel doesn't manage to charge serious money for their compiler because it > is better on some synthetic benchmark but worse in the real world. Maybe > GCC has mostly caught up with 4.3, but there's no doubt that ICC has > traditionally generated faster code. I don't have extensive/varied personal > experience with ICC, but if you compile Squeak with it on Intel, you'll see > a 20% speed-up on macro benchmarks. > > If you > want to run your software only on Intel chips, no AMD, no Motorola, no > Sun, no .... > > First, it's simply not true that Intel's compiler doesn't work for AMD. > After surfing around for 15min or so, I learned: > - at various times in the past, AMD chose ICC as the compiler they used to > generate SPEC benchmarks. > - there are numerous personal accounts where ICC generates the fastest code > for AMD for their pet application. > > Typically, Intel CPUs have a higher percentage improvement by using ICC > instead of GCC, but AMD CPUs also benefit. That's funny because AMD accuses Intel of checking for CPUID: http://www.amd.com/us-en/assets/content_type/DownloadableAssets/AMD-Intel_Full_Complaint.pdf page 40 and 41 Cheers Philippe |
2008/9/27 Joshua Gargus [hidden email]:Philippe Marschall wrote: 2008/9/26, Jason Johnson [hidden email]: On Thu, Sep 25, 2008 at 9:15 PM, Igor Stasenko [hidden email] wrote: 2008/9/25 Chris Kassopulo [hidden email]: http://www.lesser-software.com/en/content/products/lswvst/lswvst.htm It would be nice to look at it (LSWV). So many cool features. Too bad its proprietary. :( -- Best regards, Igor Stasenko AKA sig. Is the cost prohibitive? Dolphin wasn't free, but it was really cheap and a quality product. Keep in mind that while programming languages have been largely "commoditized" the best implementations of many languages still cost money. E.g. you want the best C compiler? It's not GCC. Buy the Intel compiler and watch your code run twice as fast. Let me preface this by saying that I support free software, and use it in preference to proprietary software when possible. However, I disagree with much of what you say. If you define best best as "longest bar in some benchmark". I define best as "producing the best performance for the program that I am interested in compiling". Intel doesn't manage to charge serious money for their compiler because it is better on some synthetic benchmark but worse in the real world. Maybe GCC has mostly caught up with 4.3, but there's no doubt that ICC has traditionally generated faster code. I don't have extensive/varied personal experience with ICC, but if you compile Squeak with it on Intel, you'll see a 20% speed-up on macro benchmarks. If you want to run your software only on Intel chips, no AMD, no Motorola, no Sun, no .... First, it's simply not true that Intel's compiler doesn't work for AMD. After surfing around for 15min or so, I learned: - at various times in the past, AMD chose ICC as the compiler they used to generate SPEC benchmarks. - there are numerous personal accounts where ICC generates the fastest code for AMD for their pet application. Typically, Intel CPUs have a higher percentage improvement by using ICC instead of GCC, but AMD CPUs also benefit.That's funny because AMD accuses Intel of checking for CPUID: http://www.amd.com/us-en/assets/content_type/DownloadableAssets/AMD-Intel_Full_Complaint.pdf page 40 and 41 Thanks for the link, I forgot about those allegations. Now that I have the search term "amd cpuid intel compiler", I find plenty of instances of people claiming to have independently discovered CPUID checks in code generated by ICC. Given some of the things I've read (ICC generating ridiculously slow memcpys for AMD), I wonder how ICC can still be competitive with top compilers in HPC benchmarks like: http://www.cse.scitech.ac.uk/disco/Benchmarks/Opteron_compilers.pdf It is easier to find anecdotes on less-official-looking pages like the following... http://www.altechnative.net/e107_plugins/content/content.php?content.6 ... where ICC does well. Nevertheless, even though ICC does well on AMD, it could do better without the CPUID checks, and I hope that Intel had to pay AMD a lot of money for their actions. What was the result of AMD's litigation? Cheers, Josh Cheers Philippe |
In reply to this post by Philippe Marschall
Philippe Marschall wrote:
2008/9/27 Joshua Gargus [hidden email]:Philippe Marschall wrote: 2008/9/26, Jason Johnson [hidden email]: On Thu, Sep 25, 2008 at 9:15 PM, Igor Stasenko [hidden email] wrote: 2008/9/25 Chris Kassopulo [hidden email]: http://www.lesser-software.com/en/content/products/lswvst/lswvst.htm It would be nice to look at it (LSWV). So many cool features. Too bad its proprietary. :( -- Best regards, Igor Stasenko AKA sig. Is the cost prohibitive? Dolphin wasn't free, but it was really cheap and a quality product. Keep in mind that while programming languages have been largely "commoditized" the best implementations of many languages still cost money. E.g. you want the best C compiler? It's not GCC. Buy the Intel compiler and watch your code run twice as fast. Let me preface this by saying that I support free software, and use it in preference to proprietary software when possible. However, I disagree with much of what you say. If you define best best as "longest bar in some benchmark". I define best as "producing the best performance for the program that I am interested in compiling". Intel doesn't manage to charge serious money for their compiler because it is better on some synthetic benchmark but worse in the real world. Maybe GCC has mostly caught up with 4.3, but there's no doubt that ICC has traditionally generated faster code. I don't have extensive/varied personal experience with ICC, but if you compile Squeak with it on Intel, you'll see a 20% speed-up on macro benchmarks. If you want to run your software only on Intel chips, no AMD, no Motorola, no Sun, no .... First, it's simply not true that Intel's compiler doesn't work for AMD. After surfing around for 15min or so, I learned: - at various times in the past, AMD chose ICC as the compiler they used to generate SPEC benchmarks. - there are numerous personal accounts where ICC generates the fastest code for AMD for their pet application. Typically, Intel CPUs have a higher percentage improvement by using ICC instead of GCC, but AMD CPUs also benefit. Also, many applications don't target PowerPC, Sparc, or ARM. For desktop apps, x86 is the name of the game. I don't feel limited by using an x86-only compiler, especially since I can write my code to compile with a different compiler for other platforms. If you want to put up with all kinds of trouble and issues compiling software that was written with GCC in mind like all the GNU/Linux software. That's an interesting angle. Many Free software lovers would consider code that uses specific features of Microsoft's C compiler to be broken. I generally agree with them; unless there is a good reason for it (and there sometimes is, just as it sometimes makes sense to write x86 assembly), compiler-specific code is a bug. Why should it be any different for GNU-specific code?Such things happen automatically if you compile with only one compiler. Right, but that doesn't make it a good thing. We do agree that it's not a good thing to accidentally make choices that limit Squeak to a single compiler, don't we? If not, why is taking choice away a good thing? You can observe it every time a new GCC version comes out that is more standard conformant a lot of software breaks. We're not talking about the steady improvements in standard C/C++ compliance. We're talking about intentional, compiler-specific extensions that are not part of the C/C++ standards. That's why it's called "gnuify" not "standardize" :-). You can also see it with Squeak. A lot of code has empty statements and temps declared in a too wide scope. Once we get a new compiler that supports closures or you compile it with a more standard conformant compiler, that code has to be updated. I'm not sure what we're talking about any more. I initially responded to a message from you that seemed very dismissive of Jason's point, namely that in particular circumstances, it may make sense to pay for your language implementation. For some reason, I felt compelled to argue that despite your somewhat glib dismissal, there exist real situations where it makes sense to use ICC (therefore lending support to Jason's more general claim). Given this context (which is of course not necessarily what the conversation meant to you, subjectively), what are we talking about here? :-) To me, it seems like FUD to disparage a compiler for inability to compile non-standard code. Maybe even Squeak thanks to gnuify. No, gnuify transforms the code to make it run faster on GCC, but if you don't apply gnuify, it will compile fine with other compilers.But slower. Squeak compiled with GCC is faster when gnuify is used. But Squeak compiled with ICC is faster still. On what do you base your assertion that it is slower? You don't actually pay for this, do you? For my personal Squeaking, of course I wouldn't pay for ICC. But if someone is working for a company whose product is performance sensitive, why wouldn't they spend a few hundred dollars to improve performance as much as $10000s of engineering effort would, especially if it reduces time-to-market? Cheers, Josh Cheers Philippe |
On Sat, Sep 27, 2008 at 9:48 PM, Joshua Gargus <[hidden email]> wrote:
> For my personal Squeaking, of course I wouldn't pay for ICC. But if someone > is working for a company whose product is performance sensitive, why > wouldn't they spend a few hundred dollars to improve performance as much as > $10000s of engineering effort would, especially if it reduces > time-to-market? Maybe I missed this, but - is there anything stopping someone from freely distributing binaries of Squeak that have been compiled with ICC? Avi |
Avi Bryant wrote:
IANAL, but no.On Sat, Sep 27, 2008 at 9:48 PM, Joshua Gargus [hidden email] wrote:For my personal Squeaking, of course I wouldn't pay for ICC. But if someone is working for a company whose product is performance sensitive, why wouldn't they spend a few hundred dollars to improve performance as much as $10000s of engineering effort would, especially if it reduces time-to-market?Maybe I missed this, but - is there anything stopping someone from freely distributing binaries of Squeak that have been compiled with ICC? Here are the Intel tool page and the license agreement: http://www.intel.com/cd/software/products/asmo-na/eng/index.htm http://www.intel.com/cd/software/products/asmo-na/eng/346084.htm The license doesn't talk about your code, only the materials that you obtained from Intel. An exception is the non-commercial license, so if Squeak.org decided to provide an ICC-compiled binary, we'd either have to: - clearly state that it's not for commercial use, or - cough up the money for licenses for the VM maintainers Josh Avi |
In reply to this post by Joshua Gargus-2
2008/9/28 Joshua Gargus <[hidden email]>:
> Philippe Marschall wrote: > > 2008/9/27 Joshua Gargus <[hidden email]>: > > > Philippe Marschall wrote: > > 2008/9/26, Jason Johnson <[hidden email]>: > > > On Thu, Sep 25, 2008 at 9:15 PM, Igor Stasenko <[hidden email]> wrote: > > > 2008/9/25 Chris Kassopulo <[hidden email]>: > > > http://www.lesser-software.com/en/content/products/lswvst/lswvst.htm > > > > > It would be nice to look at it (LSWV). > So many cool features. Too bad its proprietary. :( > > -- > Best regards, > Igor Stasenko AKA sig. > > > Is the cost prohibitive? Dolphin wasn't free, but it was really cheap > and a quality product. Keep in mind that while programming languages > have been largely "commoditized" the best implementations of many > languages still cost money. E.g. you want the best C compiler? It's > not GCC. Buy the Intel compiler and watch your code run twice as > fast. > > > > > Let me preface this by saying that I support free software, and use it in > preference to proprietary software when possible. However, I disagree with > much of what you say. > > If you define best best as "longest bar in some benchmark". > > I define best as "producing the best performance for the program that I am > interested in compiling". > > Intel doesn't manage to charge serious money for their compiler because it > is better on some synthetic benchmark but worse in the real world. Maybe > GCC has mostly caught up with 4.3, but there's no doubt that ICC has > traditionally generated faster code. I don't have extensive/varied personal > experience with ICC, but if you compile Squeak with it on Intel, you'll see > a 20% speed-up on macro benchmarks. > > If you > want to run your software only on Intel chips, no AMD, no Motorola, no > Sun, no .... > > First, it's simply not true that Intel's compiler doesn't work for AMD. > After surfing around for 15min or so, I learned: > - at various times in the past, AMD chose ICC as the compiler they used to > generate SPEC benchmarks. > - there are numerous personal accounts where ICC generates the fastest code > for AMD for their pet application. > > Typically, Intel CPUs have a higher percentage improvement by using ICC > instead of GCC, but AMD CPUs also benefit. > > > That's funny because AMD accuses Intel of checking for CPUID: > > http://www.amd.com/us-en/assets/content_type/DownloadableAssets/AMD-Intel_Full_Complaint.pdf > > page 40 and 41 > > > Thanks for the link, I forgot about those allegations. Now that I have the > search term "amd cpuid intel compiler", I find plenty of instances of people > claiming to have independently discovered CPUID checks in code generated by > ICC. > > Given some of the things I've read (ICC generating ridiculously slow memcpys > for AMD), I wonder how ICC can still be competitive with top compilers in > HPC benchmarks like: > http://www.cse.scitech.ac.uk/disco/Benchmarks/Opteron_compilers.pdf > > It is easier to find anecdotes on less-official-looking pages like the > following... > http://www.altechnative.net/e107_plugins/content/content.php?content.6 > ... where ICC does well. > > Nevertheless, even though ICC does well on AMD, it could do better without > the CPUID checks, and I hope that Intel had to pay AMD a lot of money for > their actions. What was the result of AMD's litigation? Not yet in court. Such things take time. Cheers Philippe |
In reply to this post by Joshua Gargus-2
2008/9/28 Joshua Gargus <[hidden email]>:
> Philippe Marschall wrote: > > 2008/9/27 Joshua Gargus <[hidden email]>: > > > Philippe Marschall wrote: > > 2008/9/26, Jason Johnson <[hidden email]>: > > > On Thu, Sep 25, 2008 at 9:15 PM, Igor Stasenko <[hidden email]> wrote: > > > 2008/9/25 Chris Kassopulo <[hidden email]>: > > > http://www.lesser-software.com/en/content/products/lswvst/lswvst.htm > > > > > It would be nice to look at it (LSWV). > So many cool features. Too bad its proprietary. :( > > -- > Best regards, > Igor Stasenko AKA sig. > > > Is the cost prohibitive? Dolphin wasn't free, but it was really cheap > and a quality product. Keep in mind that while programming languages > have been largely "commoditized" the best implementations of many > languages still cost money. E.g. you want the best C compiler? It's > not GCC. Buy the Intel compiler and watch your code run twice as > fast. > > > > > Let me preface this by saying that I support free software, and use it in > preference to proprietary software when possible. However, I disagree with > much of what you say. > > If you define best best as "longest bar in some benchmark". > > I define best as "producing the best performance for the program that I am > interested in compiling". > > Intel doesn't manage to charge serious money for their compiler because it > is better on some synthetic benchmark but worse in the real world. Maybe > GCC has mostly caught up with 4.3, but there's no doubt that ICC has > traditionally generated faster code. I don't have extensive/varied personal > experience with ICC, but if you compile Squeak with it on Intel, you'll see > a 20% speed-up on macro benchmarks. > > If you > want to run your software only on Intel chips, no AMD, no Motorola, no > Sun, no .... > > First, it's simply not true that Intel's compiler doesn't work for AMD. > After surfing around for 15min or so, I learned: > - at various times in the past, AMD chose ICC as the compiler they used to > generate SPEC benchmarks. > - there are numerous personal accounts where ICC generates the fastest code > for AMD for their pet application. > > Typically, Intel CPUs have a higher percentage improvement by using ICC > instead of GCC, but AMD CPUs also benefit. > > Also, many applications don't target PowerPC, Sparc, or ARM. For desktop > apps, x86 is the name of the game. I don't feel limited by using an > x86-only compiler, especially since I can write my code to compile with a > different compiler for other platforms. > > If you want to put up with all kinds of trouble and > issues compiling software that was written with GCC in mind like all > the GNU/Linux software. > > That's an interesting angle. Many Free software lovers would consider code > that uses specific features of Microsoft's C compiler to be broken. I > generally agree with them; unless there is a good reason for it (and there > sometimes is, just as it sometimes makes sense to write x86 assembly), > compiler-specific code is a bug. Why should it be any different for > GNU-specific code? > > > Such things happen automatically if you compile with only one > compiler. > > Right, but that doesn't make it a good thing. We do agree that it's not a > good thing to accidentally make choices that limit Squeak to a single > compiler, don't we? If not, why is taking choice away a good thing? > > You can observe it every time a new GCC version comes out > that is more standard conformant a lot of software breaks. > > We're not talking about the steady improvements in standard C/C++ > compliance. We're talking about intentional, compiler-specific extensions > that are not part of the C/C++ standards. That's why it's called "gnuify" > not "standardize" :-). I was talking about unintentional use of of compiler-specific extensions that are not part of the C/C++ standards or reliance on implementation specific behavior. That happens a lot and often results in breakage in never GCC releases. For example the current vitualbox does not compile with the current release of GCC (4.3) and qemu does not build with any version 4 of GCC. > You can > also see it with Squeak. A lot of code has empty statements and temps > declared in a too wide scope. Once we get a new compiler that supports > closures or you compile it with a more standard conformant compiler, > that code has to be updated. > > > I'm not sure what we're talking about any more. I initially responded to a > message from you that seemed very dismissive of Jason's point, namely that > in particular circumstances, it may make sense to pay for your language > implementation. For some reason, I felt compelled to argue that despite > your somewhat glib dismissal, there exist real situations where it makes > sense to use ICC (therefore lending support to Jason's more general claim). Suddenly we have a much more more differentiated statement that it's much easier to a agree on than "ICC is the best compiler, it makes your code run twice as fast". Cheers Philippe |
Free forum by Nabble | Edit this page |