Slashdot Mirror
HP Introduces Final Processor in PA-RISC Family
The HP Way writes "According to an article on InformationWeek, HP announced the immediate availability of the 800 MHz, 1.0 GHz, and 1.1 GHz dual-core PA-8900 with 64MB on die L2 cache, the last member of the PA-RISC family of microprocessors. Customers with Superdome chassis can install Itanium 2 CPUs alongside PA-8900 processors."
Absolutely not. The new macs are x86, and only x86 not itanium. Read the info from apple.
jeff
ipv6 is my vpn
Wrong!
Compaq bought Tandem, switched them to Alpha (was mips) and now HP is moving them to Itanium.
Tandem has never used PA-RISC
Well I might be wrong here but I think that HP helped Intel design the Itanium. They've been planning this for several years I think
I myself do not understand the purpose of the x86 cruft any longer. Nostalgia? Are people buying Pentium 4s to run DOS in Real Mode?
Not since Marie-Antoinette played milkmaid has looking simple and honest been so fake and complicated.
As I explained, it just doesn't matter which one is "worst". Most modern CPUs have a user-visible "skin" slapped over some exotic out-of-order set of execution units. Your view of what's better or worse is just a superficial impression of what the skin looks like.
If all of those designs truly had more potential than a design with an x86 skin, then at some point one of them would have permanently pulled ahead in performance - even if only in the server arena. This just hasn't happened.
The PPC in consoles probably has more to do with IBM's scheme to conveniently include a few DSPs on the die than with any deep architectural differences in the main Power core. It's a just packaging optimization targeted at the embedded media appliance market.
What you say about IBM and H1B workers isn't true; I've worked for IBM as an H1B worker yet I do not have a Ph.D, and many of my colleagues on the project we were on were also on H1B workers. There was a critical (and genuinely rare) piece of experience we all had, but other than that we were just normal engineers.
Additionally, I was paid significantly *more* than the native IBMers because they paid me an International Service Allowance (which was generous enough I could live off it and spend hardly any of my actual salary) - so IBM was certainly not abusing the H1B system to hire cheap foreign workers because none of us were cheap.
Oolite: Elite-like game. For Mac, Linux and Windows
Are these CPU's REALLY that good in the end? I mean, if we look at this particular CPU: It has 64MB of L2-cache. Now, is this really a kick-ass CPU, or is it a mediocre CPU that hides it's crappiness behind lots and lots of cache?
RISC processors typically run more efficiently with more cache than their CISC counterparts because of larger instruction sizes. The typical applications that the PA-RISC is targetted toward (large databases, high-end cad and engineering work, and so on) have larger working sets than the typical applications of an x86 (web serving, games).
However, this particular product is clearly aimed at current PA-8700 and PA-8800 users who are unwilling to switch platforms yet. There is little incentive for someone to switch from Opteron or EM64T to the PA-8900 as integer and floating point performance leave a little to be desired. Like the release of the Alpha EV7z, no serious effort was made to improve the performance of this design. As with the EV7z, the maximum clock frequency was decreased from 1.6 GHz to 1.1 GHz. Pretty much all of HP's semiconductor design and testing staff has been sacked or transferred over to Intel.
In reality, the PA-8900 is more of an advertisement for the Itanium 2 than a serious attempt at microprocessor.
Sorry, you are wrong and I spoke with Patterson just a few days ago at the Internation Symposium of Computer Architecture in Madison, WI.
Patterson and Hennessy argued for RISC in the 80s before technology allowed Intel and AMD to burn 3 million transistors on a CISC->RISC translation layer. They did not forsee x86 hanging on until the mid-90s to enable this. So yes, they are wrong about the death of x86 but modern out-of-order superscalar pipelines are all based on the principles of the early RISC 5-state pipelines.
But your post claims they are failures and you are dead wrong. Among numerous other contributions, you can thank Patterson (and Randy Katz) for RAID.
Seriously, the first H+P textbook shaped the way a generation of computer-architecture students think about the subject, surely including some of the x86 designers who have done such an admirable job over the last decade. Of course, some of the particular architecture ideas of the MiPS and RISC projects turned out to be short-lived, but the general lessons have been well absorbed.
You are misinformed. I personally know at least two foreign (non-Japanese) Fujitsu silicon design engineers working on American soil.
J
In the server market, only 2 RISC chips remain. They are the PowerPC by IBM...
Outside of the Apple Xserve and a couple BladeCenter line machines, where has PowerPC ever been used in servers?
Perhaps you meant POWER (Performance Optimization With Enhanced RISC)? Largely unrelated architecture. For what it's worth, POWER5 is actually doing extremely well. They continue to have extremely high performance and scalability, and with the Blue Gene project, will probably be used in the world's fastest computers for some time.
here is probably the best source saying what will be in the new Intel based Macs. it's the universal programming guidelines from Apple. it states that the instruction set developers should use revolves around x86, not x86-64. it would be stupid for Apple to tell developers one thing now, then change it again a year from now when Apple releases the consumer Intel based Macs. they are trying to make it easy for developers to support both PPC and Intel based Macs. the easiest way to do this is to have something in common, and both have 32-bit lineage.
This also gives us a good look at what will be in the first Intel based Macs. It will not be a 64-bit chip. If it were, Apple would tell the developers they are free to program for that architecture because that's what all the new Macs would be based on. Since Apple isn't saying to use x86-64 instructions, we can at least assume that some of the Intel based Macs aren't 64-bit. It would be easier on the developers to support the Intel side. Eventually, though, Apple will no doubt use the 64-bit chips because the x86 is getting a bit stale.
hackers of the world unite!
That they did, in fact the Itanium is able to transcode PA-RISC instructions so most PA-RISC code will run unmodified on Itanium machines.
Well that Apple engineer apparently missed the documentation that specifically states than Mac/x86 won't use OpenFirmware. Either that or you just made that up.
Seeing as there isn't a single piece of developer documentation mentioning IA-64, I think it's pretty safe to assume that none of Apple's hardware will be shipping with Itanium. You know, because they wouldn't be compatible with all of that software they're trying to convince Mac developers to port to the x86 right now. If they intended to use three ISAs they would have included IA-64 now so software can be ported to it.
Not to mention that Intanium 2 units are hot, have a weak GCC backend, and are incredibly expensive.
POWER isn't "largely unrelated" to the PowerPC. As the name suggests, the PowerPC is a PC oriented version of the POWER processor. The architecture is IDENTICAL as are NEARLY ALL of the instructions. The PowerPC is just a cost-reduced version of the POWER processor meant for markets that can't afford the monster POWER is today.