Slashdot Mirror
Sun's Zippy New Chips
Mark the Revelator writes: "Reuters has a story about Sun unveiling it's latest and greatest UltraSparcIII chips. The new chips are being made by TI and are the first UltraSparcs to use copper instead of aluminum for transistor connections. Although they're supposed to compete with Intel's Itanium chips, they only run at 900MHz ... for now."
But Britney does have talent!
It's just that you usually the only place you see that kind of talent is in porno movies...
If it's supposed to move and doesn't, use WD-40. If it moves and it shouldn't, use duct tape.
Ars Technica has a fantastic article comparing and contrasting Sun's future MAJC (Microprocessor Architecture for Java Computing) CPU architecture with Intel's IA-64. It's going to be very interesting to see if Sun can carve out a large enough market to ensure MAJC's viability. My uninformed opinion after reading the article--Sun has been making decisions since its founding that have given it the only chance to survive. By almost totally eschewing both Intel and Microsoft, Sun has been forced to innovate on both hardware and software to compete with these giants. Sun simply had to invent Java--what was the alternative, reselling NT "workstations"?! Now Sun has leveraged Java into strategic partnerships with IBM, Oracle, etc. to create from scratch a major software niche, not to mention Java's future in the embedded markets. MAJC it seems to me is the logical step in hardware once Sun made the commitment to Java and once Sun decided not to become a reseller of Intel chips like say HP. Without having to worry about what Intel wants, Sun can use its traditional RISC approach to registers to once again offer a fantastic alternative--read the Ars Technica article cited above: "MAJC, however, spends so much of its die space on registers that it can have the register states for four different threads loaded at once. Since it doesn't have to save and load register states to switch between threads, its context switches are very fast". In the 1980s HP saved the company investing in PA-RISC. Maybe that was because the engineer founders David Packard and Bill Hewlett were still alive and strong. I believe that it is Sun that has applied that lesson of not surrendering control over the CPU architecture, and that HP will continue to pay a heavy price for deciding to go with Intel. Financing new chip architectures is difficult, but in my opinion there is no future for being a reseller unless one is IBM or Dell. (And note that IBM resells only because it wants to since it already manufactures alternatives, it is beholden to no one. Just who will be able to compete with IBM's Global Services?)
I don't know why you buy computers for, but we buy them to run our applications. I don't give a flying f*#& about benchmarks, CPU speed, Cache and any other dribble PC manufactures or reviewers or dipshits like yourself spew about. What I care about is getting the job done as fast as possible and as cheap as possible. For MY work and MY applications that solve MY problems, Intel machines running Linux perform at a cost performance ratio 6x to 8x better than anything SUN makes. YOUR milage may vary. Your intelligence or lack thereof will help keep SUN around. Go ahead an flush your $$$ away. You've obviously flushed your brains away.
"TV, a medium as it is neither rare nor well done." Ernie Kovacs
Future processors will, at least partly, be reconfigurable, that is they will load a set of instructions and save it, and then have to load the data only.
You just described a cache, and they have been used for years.
Oops, sholud be this
The report only mentions SSL and the numbers 331 for Itanium and 32 for US2. They ran different software on different operating systems on different hardware platforms. Do you understand what that means? Did they produce a random number ever, if so how? Solaris doesn't come with /dev/random or /dev/urandom, but Linux does. OpenSSH runs around 60 commands and weighs the times if /dev/urandom isn't available to read from. What does the software coradiant use do?
TI has been the major supplier of Sun's Sparc chips for a long time.
I realize that, but the sentence "Although they're supposed to compete with Intel's Itanium chips, they only run at 900MHz" implies that the Itanium runs faster than 900mhz... that is what i was referring to. I didn't really get the point of that sentence since they are both 64-bit chips, and if anything, the Sparc should be faster clock-for-clock than the Itanium.
-Colin
Based on what data? Are they faster all around... in every type of computation? I doubt it. I'm sure they are much faster at some types of instructions, and I'd take a Sun over an Intel machine any day, but the performance depends on what types of programs you're running (and how many people are going to be running them at the same time.)
"The best laid plans of mice and men gang oft agley..." - ROBERT BURNS
Yeah, but only the kind with feathers.
There is much cruelty in the universe, John.
Yeah, we seem to have the tour map.
Right. EPIC is a little bit different than VLIW. With VLIW, the compiler has to know the number of execution units on the processor at *compile* time. This isn't a problem for custom embedded systems, but for shrink-wrapped software it is a *big* problem. Why? Because when a new processor is introduced with more execution units, the old binary code won't run on it. EPIC makes code that doesn't depend on the number of execution units.
Tim
Intel D850GB motherboard Pentium 4 1.7 GHz: SPECint2000 = 586
All from http://www.specbench.org.
My old manager had a great reason for buying Sun; no one will ever put M$ windows on it!
...Solaris went out of business and was taken over by Sun...
Can you fucking dumbass college students at least sober up before posting your pro x86 dribble that you've been brainwashed into thinking? Nobody gives a flying fuck that you smoked bowl for breakfast.
P.S. I now have "only" 49 karma points. You've injured me to the core! hehehe
Although they're supposed to compete with Intel's Itanium chips, they only run at 900MHz ... for now."
Itanium only run at 733 or 800MHz
IBM RS/6000 POWERserver 590
My webserver, in fact. (though it's done other things as well)
In general when comparing archtectures, then yes, some architectures are better than others at some things... but the x86 is CRAP AT EVERYTHING. IT SUCKS. IT HAS ALWAYS SUCKED.
;)
>>>>>>>>>
I wouldn't go that far. It sound like trolling to me.
It is register starved, has a crappy "grown" CISC instruction set that's a bitch to optimise,
>>>>>
I don't know about GCC, but Intel does a damn good job at it. (Check out Intel C++ 5.0)
has a wierd memory map that, again, is a bitch to optimise, tends to have crappy I/O throughput, cache coherency issues (related to aforementioned instrcution set and memory map),
>>>>>>
What "weird" memory map? I've actually read the whole x86 system designer docs, and the only wierdness is that I/O devices are mapped in at 640K instead of high up where they should be. That and the 4GB virtual address space (think big mmaps) but you can hardly fault it for being 32 bit.
highish latency on task switching
>>>>>>
Only if you use the standard TSS system (which nobody does). Plus, it has lots of optimizations (like global PTEs and delayed saving of the FP/SSE registers) that make task-switching quite speedy. Plus, from the (admittedly old) lmbench benchmarks, Linux 1.3 on x86 context switches faster than any other OS on any other platform. (www.bitmover.com/lmbench, I think). Plus, QNX manages to get. 4.5 microsecond task switches on Pentium-class processors, so I doubt x86 task switching is THAT slow.
(well... that depends on how enamoured you are of memory protection, of course - see AmigaOS/AROS on x86 for near-minimal latenncy, with no mem-protection...) and is JUST CRAP.
>>>>>>>>>
QNX does memory protection (as does Linux
Basically, for a given MHz rating, the x86 will always come out slowest, no matter what test you do. Except running x86 binaries, of course...
>>>>>>
Actually, the P4 runs SSE ops *really* fast.
Thanks to Open Source, I can now happily use Linux on PPC and ARM boxes, and there's finally some nice, cheap PPC MoBos that aren't apple coming onto the market.
>>>>>>>
So you want to pay more to get less functionality?
Seriously, x86 doesn't suck as much as you think it does. Intel and AMD have invested a huge amount of money in improving the architecture, and even if it is often slower in instructions per clock, the sheer clockspeed of x86 chips often make them outperform their competitors. Plus, they have great price/performance ratio.
A deep unwavering belief is a sure sign you're missing something...
One of the reasons the Intel universe gets a bad rep for reliability is the cheap hardware people cobble together and call a "server."
Some people think a machine with 256MB of SDRAM and two 30GB IDE drives make for a good server. Never mind considerations of ECC memory or mirrored disk. Let's not even talk about using overclocking in a server!
I've seen some pretty good reliability from properly architected Intel boxes. As good as a real Unix server? Not yet, but they're getting ever closer.
Amen.
I have a Sparc IPC, Sparc5 and an Ultra 1. Never have I seen any hardware glitch. The only software problems have been caused somewhere between my chair and my keyboard.
Well, I've been happily using SUCKY computers since 1983. There's nothing more satisfying than using a really good kludge. Running a direct descendant of the world's first microprocessor (the Intel 4004) at thousands of times the original performance level is an awesome kludge. Throw in all of the goodies availible for x86 machines (OS's, apps, hardware) and its been like spending 18 years in a candy store.
You can stay up there in your ivory tower and watch the rest of us having fun down here in the fresh air.
Even similar architectures can perform differently at the same clock speed. Look at the PIII and the P4. IIRC, a P4 at 1.7MHz is approximately as fast as a PIII at 1.3 MHz.
"The Artist, seeking Beauty, discovers Truth; The Scientist, seeking Truth, discovers Beauty."
I don't expect an UltraSparc comparison, but given that there are known performance issues in the P4 with "legacy" code, I wonder how easily that benchmark could have been stacked. How well would it compare with a PIII?
Or if you want a machine that was designed and built by engineers, not by a 14-year-old in his basement (or the Dell equivalent thereof).
I'm not a huge fan of Solaris, but Sun hardware is good stuff. I think SGI hardware is better stuff, or used to be before SGI decided it would be fun to implode.
You buy Sun if you want scalability and reliability. You buy Intel if money is an issue.
Potato chips are a by-yourself food.
Well, this is old news! Actually, SUN couldn't get TI to manufacture the cheap in high yields at 900mhz. This is why most of their new machines are coming out with 750mhz instead of 900mhz. Moreover, SUN said they would switch to a Taiwan based manufacturer to get the yields up for 900mhz and above so that they don't fall behind Intel/AMD in the MHZ game!
Know what makes me feel old? My first box ran at 4.77MHz. Yeah. 4.77. Umm. And it had two floppies! RaH! No hard drive, just two floppies.
And later I got a 300 baud modem. Double RaH.
Later,
daWiz
The Wizard's Realm - Telegard 2.0 - 686-0235
Here's a real Zippy processor.
"Rocky Rococo, at your cervix!"
The new chips are being made by TI
Does this mean that I'll be able to run the "Space Rocks" game I programmed on my TI-85 graphing calculator?
My religion forbids the use of sigs.
Since I've been working in an HP shop I've noticed HP seems to be the most uninteresting tech company. With the exception of the Superdome they haven't done anything innovative in a decade. Maybe I'm just bitter, but HP seems to be a "me too" company.
An even better analogy than the F-1 in traffic is doing a straight comparison to a pickup.
Give the racer and the pickup each a load to haul. The racing car will pass the pickup several times hauling the load, but the truck driver will be twiddling his thumbs waiting for the other guy to finish.
While Sun boxen are only decent at CPU power, that was not the central design goal. Pushing data is.
-RobHood
I'm not an anti-{insert OS} zealot. I just like blowing people's little minds.
This is about the closest I've seen to a multi-G4 machine: www.totalimpact.com. They have 4 and 8 node G4 clusters. Each CPU node is smaller than Apple's Cube, and the 8 node cluster (expensive at $28k) weighs in at a mere 22lbs..
No. Sun's chips have always been fabbed by TI. I believe, however, that TI just recently started fabbing chips with CU interconnect.
Duh!
Your Servant, B. Baggins
Durham University USIII Cluster running Gridengine went live yesterday (31st June). It's now the highest performing academic supercomputer in the UK.
You can't compare a P4 with UIII. The SUN processor is 64bits and costs alot more. I don't want to see a P4 in a server.
Well there is the Blade 1000, which uses the US3, but that box is far, far from cheap with a price tag of around $9000 USD.
:P
But then again, your boss really likes you and you know you can con him into buying you hardware necessary do get your work done
I recently received the sunfire 280r I ordered for use as a network management station. One of the first things I noticed was how fast it was able to un-gzip files. It sounds silly, but I didn't think it worked the first time I ran the command, the prompt just came back too fast. I thought, "hey, what the hells wrong with gzip?", then realized that it was just a lot faster than the other systems I had been using.
I know its just anectdotal, but hey, some of the best stuff is!
Casca
The P4 should have shown everyone that MHz dont't say anything. In most cases, a fast Athlon or P3 can beat the almost 400MHz faster P4. So there's no reason why a 900MHz Sparc or a 866MHz G4 shouldn't be able to compete with the CPUs above. MHz don't count - it's that simple.
Won't somebody PLEASE think of the child processes?!?!
--
"Outlook not so good." That magic 8-ball knows everything! I'll ask about Exchange Server next.
An F1 stuck in down-town NYC would certainly attract more chicks than an Accord in Montana. Isn't that why we choose the flashiest hardware we can, to get more chicks? :-)
:-)
On the other hand, a sparc runs the software I want to run, and the software I earn tons of money from. So of course, having tons of money gets higher quality chicks better than any car
the AC
[not a politically correct post since I'm in a country which has outlawed 'Merkin correctness]
Hemos is like...sci-fi fans;he thinks technology is cool, but he hasn't bothered to understand the science it's based on
This was an interesting remark, I started to hear similar things very often recently. My understanding is that most EDA software on the x86 platform is written for WindowsNT/2000. Isn't your productivity to some degree limited by the stability of the Windows operating system, especially for CPU-intensive, lengthy tasks like formal verification and transistor-level simulation? (But then, these things are most likely done on server farms in big companies, anyway) Are you using your x86 machines for these tasks; or are you mostly using them for HDL design entry/simulation; and synthesis?(which will admittedly be a lot more cost-effective on x86 systems)
It's just sad that many EDA companies tried offering their products on Linux; and had to pull them back because of low demand.
Zigbee Central: A Zigbee weblog
Don't get me wrong, I love my Athlons, but I used to work in an HP based shop with PA-RISC all around. I'll never forget when the K-Class and N Class servers first came into our data center with the latest PA-RISC beasts - they were so fast it was scary (this was like 3 years ago)
Top Most Bizarre/Disturbing Error Messages
This *really* depends on what your doing. I mean if you run Seti (or things with massive FP) then a US450 is about the equivalent of a 900Mhz Pentium III. (I have did many tests to prove this). BUT if you talking integer performance (with the exception of RDBMS) then the Pentium III is faster. I just got a Sunfire 280 in (2x750USIII) and I can't wait to play around with it.
Space may be the final frontier, but it's made in a Hollywood basement. --Red Hot Chili Peppers, Californication
starting at 10,000 bucks a throw i should think so!! (monitor not included) http://store.sun.com/catalog/doc/BrowsePage.jhtml? catid=55877
Hmm, Not really.
I mean, yeah, they're totally different. And they're faster clock-per-clock (with added benefit to FP stuff).
But a 1.4GHz Athlon blows away a 7-800MHz UltraII for most kinds of computation. A 1 GHz Athlon seems to be about (42, 29) on the (retired) SPECint95/SPECfp95. A 450Mhz Ultra-II (not Ultra-IIi, I'm looking at results for an SPARCstation Ultra-60) gets about (20, 27). That's a bit faster int clock-per-clock, and a lot faster FP. Note that for practical stuff (databases, web, whatever) int is more important. Of course benchmarks are hard to interpret, but this gives you an idea. All the SPEC benchmarks are available at www.specbench.org. Of course there are no Ultra-III results, but I'm guessing it's not going to be 2x as fast as the best x86s (at least I'll wait to see the results before I believe it).
You use a Sun because you want an architecture that will scale smoothly up to 64-way (I *guarantee* that will be faster than any single x86 machine).
Actually if you want to both go fast at the low end and scale well, you can buy an RS/6000 -- IBMs Power3 and Power4 chips are absurdly fast and scale very well (and actually focus on memory bandwidth for database performance). But a bottom-of-the-line Sun is a lot cheaper than the cheapest RS/6000.
Full disclosure: I work for IBM (in software) and I've seen a good bit of internal stuff about IBM chips, esp. the upcoming Power4. Most of that information has now been published in MicroProcessor Review and is now publicly available, I think you'll find it if you poke around...
(even more amusing full disclosure: I'm a huge fan of old Sun stuff, their machines are beautifully engineered. i use a couple old 32bit sparcs for all kinds of things)
Yes, but MHz is essentially meaningless as we all know. One of the new dual-800MHz Apple Macintosh G4s would run rings around this, even if they do manage to increase the clock speed.
if you look really close you can see the TI-99/4a instruction set.
if they had delivered on time instead of 3 years late.....
Already done.
They even have a compiler plugin for VC++ that generates fast binaries. Check this out.
Space may be the final frontier, but it's made in a Hollywood basement. --Red Hot Chili Peppers, Californication
Somebody got the reference
The Itanium achieved some truely awesome SPEC-FP scores that made Sun look pretty bad. At FP, Itanium whales.
Itanium suffers from the same problems as the Pentium 4, in some ways, in that you can't ever branch. If you can find code that does this, and doesn't have many NOPs, the Itanium will perform very well. That doesn't describe much general-purpose code in the real world.
So, the crux of this is that Itaniums are faster at some things, just like the Pentium 4 is faster at some things. The risk is that these Intel processor applications are becomming highly specialized, and better general-purpose processors are available.
I think the UltraSPARC III is about 2-3 years late to the table.
The old RISC hardware chips are dying off slowly since they can't afford the build the kind of expensive Fabs that are needed to compete with Intel and AMD.
Except IBM, of course. I did read something about the Power4 (see also this pdf ) and its emphasis on maximizing memory BW - some of the figures sounded awesome. I was really looking forward to being able to pick and choose between Power4 and 21364(pdf), but Compaq seems to have throttled the Alpha. As if the IA-64 is better!
For scalability, though, I have to wonder if rack-mounted Alpha4's connected with high-speed interfaces like Myrinet could provide an alternative to hardware like a Sun ES10000. I haven't tried Scyld or MOSIX to know if they make using such clusters a "smoothly scalable" solution. The big Sun SMP machines are nice, but they're also expensive and the aging UltraSPARC II processors are nothing to write home about anymore.
"Provided by the management for your protection."
There are chicks in Montana? Plural?
- Sig this!
OK, so those raw numbers look pretty close......out of curiosity what is the difference in price? Its the price/performance that Intel usually wins with, right?
Ther're so much more to buying Sun kit than CPU cycles.
You get (IMHO) the best OS you can run on a server (*incomparably* more reliable than Linux in my experience).
You get a better build quality than with PC class gear. (not so with the low-end Ultras I know, but have you tried carrying an E450 around lately?) I've worked with Sun boxes (mail hubs, NIS servers etc.) that haven't been swtiched off in 8 or 10 years.
You get excellent support - hardware or software. It costs, but it's worth it.
You get as much SMP as you could want.
You get insane amounts of addressable RAM and faster bus speeds.
In short kids, you get *proper computers* running a *proper OS*.
Doesn't apply well at all. They're all RISC now, but with CISC wrappers on some, you decide which. BTW: The Sparc III, code name Cheetah, is designed with multi-processor scalability in mind, including local and global cache coherency. Its databus is 128bits wide and I hear from reliable sources so close, that the 1.2 Ghz is only limited by product yield.
Just because the MHz on the Sun equipment (900MHz) is lower than the current Pentium (1.5MHz), don't be fooled into thinking the Intel hardware is better. What matters after all, is throughput and pumping that data. Check your specs!
Check this 4 CPU Intel vs the 1 CPU Sun considering plain speed...
CINT2000: Intel Corporation Intel D850GB motherboard(1.5 GHz, Pentium 4 processor) - 536 524
CFP2000: Intel Corporation Intel D850GB motherboard(1.5 GHz, Pentium 4 processor) - 558 549
CINT2000: Sun Microsystems Sun Blade 1000 Model 1900 - 467 438
CFP2000: Sun Microsystems Sun Blade 1000 Model 1900 - 482 427
CINT2000: Advanced Micro Devices Tyan Thunder K7 Motherboard, 1.2GHz Athlon MP Processor - 522 495
CFP2000: Advanced Micro Devices Tyan Thunder K7 Motherboard, 1.2GHz Athlon MP Processor - 481 433
Throughput on the Sun with 2 CPU, but strangely enough, none for any Intel hardware. Throw a 2 CPU AMD in there, though...
CINT2000 rate: Sun Microsystems Sun Blade 1000 Model 2900 - 10.7 9.97
CFP2000 rate: Sun Microsystems Sun Blade 1000 Model 2900 - 10.2 9.09
CINT2000 rate: Advanced Micro Devic Tyan Thunder K7 Motherboard, 1.2GHz 2CPU - 10.8 11.1
CFP2000 rate: Advanced Micro Devic Tyan Thunder K7 Motherboard, 1.2GHz 2CPU - 8.30 9.14
"Beware of he who would deny you access to information, for in his heart, he dreams himself your master."
Yeah, but Sun has always had pathetically slow chips whether running Slowaris or Linux. Their only advantage is that new chips from Intel tend to be slow in their first offerings. But now, with Intel having purchased Alpha from Compaq, and if they realize what they have and incorporate it into future chips, Sun will remain further back with slow chips and a lackluster OS. Sun is the Microsoft of the UNIX world. Great marketing, so-so product.
Wow, I heard completly different argument about 5 years ago, regarding the reason why RISC is faster (MhZ wise) than CISC.
Uh.... No. The Alpha was a RISC design. It used simple instructions and a long pipeline so that the amount needed to be done each clock was very little. This allowed them to ramp up the clock speed. For a while, the Alpha had the highest clock speed of any processor (I remember 866Mhz Alphas while Intel was still working in the 300-400Mhz range).
The Intel is more of a mixed design---some instructions are complex, but the pipeline on Intel chips is long. As with the Alphas, very little is done per clock cycle, so they've been able to push up the clock speed. Intel has lots of $$$, so it hasn't been difficult for them to achieve very high clock speed levels.
Btw, does anyone know what the current Alpha clock speed is?
The CISC philosophy, which Sun adopts, is to have fewer clocks, but to get more done each clock. CISC designs don't need such a long pipeline and hence aren't as susceptible to branch mis-predictions. However, the higher Mhz processoros usually still win since they have long pipelines which allow many smaller operations to be done simultaneously. It might take 20 clocks to finish an instruction, but an instruction completes almost every clock. (Comp. Sci. professors like to use a "laundry" analogy.) Branches ("if," "while," "goto" in code) can be deadly, but only if the processor guesses the wrong branch to take. Fortunately for Intel and others that adopt the long-pipeline philosophy, simple algorithms can predict the correct branch with very high accuracy (99.9%+). Hence, long-pipeline chips tend to get stuff done faster than CISC-minded models.
Jason
Until a decent MP chipset is rolled out for wintel platform, sun looks pretty. Compaq is supposed to be rolling out something good, on the mp front. speed has little to do with it, and lucky for sun , intel is rationing l1 cache, crippling otherwise good processors. maybe this is a price/marketing thing like xeons. if amd rolls out its trump card, and if intel follows, the price/performance drops in a lacklustre market - so everyones profits fall - and revenue growth. Be excited about the good stuff being held back. also lowish memory prices mean premiums for top end cpus drop.
Since you apparently are not very familiar with computer architecture R&D I'll clue you in before you make an ass of yourself again.
By loading instructions, what the poster meant was that the CPU can have an area of reconfigurable gates with which to implement a function directly into hardware. After that's done, any time the CPU needs to run a certain function, it can use the hardware it on-the-fly configured exactly for that purpose and gain an enormous speed up.
keep in mind that these are pure RISC processors and have always toasted any CISC or CISC-to-RISC processor of a much higher processor rating.
...that I would never see the words "they only run at 900MHz". Okay, so I'm only 19, but I feel old now.
Here's the formula for computing a metric that allows you to compare two different chips regardless of architecture when running a specific program:
Effective Speed( Program p ) = Instructions in p * Avg # of cycles to complete 1 instruction * seconds per cycleIn other words seconds per cycle is the inverse of clock speed.
What can be gained from this equation:
Stop drooling about clock speeds, it is nearly meaningless, and is only a marketing tool. If users thought feature size was cool, we'd be having this argument about how "0.18 micron feature size" is meaningless when trumpeted out of context.
Start getting interested in SPECint and SPECfp metrics. Why don't chip makers start focusing on those metrics?
When Moore's law starts failing, someday, we'll see far more innovation in chip designs that don't relate to feature size or clock speed. There's a whole unrealized sea of optimization that could happen to speed up current designs right now. We all beat up Microsoft for being a monopoly, how about Intel :-).
------ Tim O'Brien
> keep in mind that these are pure RISC processors
... have always toasted any CISC or CISC-to-RISC" is a solid lie. There are plenty of occasions where a CISC processor outperforms a RISC processor.
> and have always toasted any CISC or CISC-to-RISC
> processor of a much higher processor rating.
That is misleading and, in fact, bordering on the level of a total lie. The benefits of RISC architectures are not performance. They're simplicity. This simplicity, in the past, sometimes had the benefit of increasing performance, but higher performance is not a rule in and of itself.
Saying that "pure RISC processors
In specfp_2000, the lowest frequency Pentium 4 scores a 516, while the highest frequency UltraSPARC III scores a 482. The slowest Pentium 4 is 7.1% faster than the fastest USPARCIII.
In specint_2000, the slowest P4 gets a 490, while the fastest Sun processor gets a 467. Here, the wimpiest current generation Intel processor is 4.9% faster than the best thing Sun can offer.
These above factors keep in mind that the Sun chips are *specifically* architected to achieve the highest performance possible, pretty much regardless of cost. They are full-on server chips. The Intel Pentium 4 series are designed with cost factors in mind. The Pentium 4 cannot be a three thousand dollar behemoth due to its target market (actually, the 750MHz USPARCIII processor module costs about $7k on Sun's website). So the USPARCIII can have the benefit of loads of added performance enhancing features while the Netburst (P4) architecture has to cut corners at every step.
the UltraSPARC III outperforms the Pentium 4 on a clock for clock basis. Of course, the original Pentium outperforms the Pentium 4 on a clock for clock basis on many benchmarks, too. This means nothing. It is merely reflective of different design strategies. I can easily point out the fact that the Pentium 4 offers higher performance per watt or higher performance per number of integer ALUs. But, like "performance per megahertz", those are also stupid measurements.
There is nothing out there which would cause me to believe that an x86 processor made with the design strategy of the UltraSPARC III ("we're gonna sell this for thousands of dollars, so throw in the kitchen sink, too!") would not outperform the UltraSPARC III at like frequency. Well, except if the the fp instructions on the USIII are three operand, but that's a special case. ^_^
-JC
http://www.jc-news.com/
We've ported our software from Sun to Linux (and to SGI, we used to have it on IBMs as well, but there is no amrket there anymore). [ http://www.jasongeo.com for those who care about cutting edge seismic inversion, stochastic inversion and other really cool stuff. Though that site is the marketing friendly version ] Anyway, we found that for our applications a 1Ghz PIII is about 3.5 times as fast as a Sun ultra 10/60/80 (with an Ultra II 440 cpu). A blade 1000/750 is about half as fast as a PIII/1000
While it may be up for some debate, CISC processors are known primarily for opcodes of variable length (some instructions are one word, some two, some three, etc.). They are also known for an overly-rich instruction set and a smaller number of registers.
RISC processors are known for uniform length of opcodes, a library of instructions that has been tuned to optimize compiler output, no direct operations on memory other than load/store, and a larger number of registers (sometimes allocated in clumps like the Sparc register windows).
VLIW (which Intel calls Epic) is also known for uniform length of opcodes, but gathers several opcodes into a bundle and executes the bundle all at once. The Itanium executes 3 instructions at a time in this way. The Itanium's main weakness is that it cannot execute these "bundles" out-of-order, and it relies upon compile-time analysis for most of its optimization when the least is known about the executible code.
AFAIK, both the Pentium 3/4 and the Athlon still have out-of-order RISC processors at the core, and translation units that move x86 code into the native code. Cyrix was the last vendor to make a real x86-CISC processor, but it couldn't scale much beyond 400MHz, so VIA killed it.
The original Pentium had something similar to the VLIW idea in that it had two parallel execution units for executing two instructions at once (super-scalar), but the second execution unit got switched off if there was a dependency between the group of two instructions.
Different architecture; the blade 100 uses 500MHz US2i chips, these are US3. They're unlikely to release a blade 100 with these chips in the near future. Longer term, I'd imagine a successor to the Ultra 5 type systems will be forthcoming at some point in the next 6 months or so, but it would almost certainly be more expensive.
> Itanium chips, they only run at 900MHz
> for now.
C'mon!! Raw clock speed is not the best-- or even a "good" --measure of a microprocessor's performance. You've been reading to much of the rhetorical effluvium that spews from the marketroids in the x86 camp.
Is this a good idea though? I mean, using one of today's compilers, ported to a IA64/Itanium architecture, a compiled program might run very slowly, since today's compilers probably let a bit of the optimization (within reason) up to the CPU. This would also mean that it may be a little while until some quality IA64 compilers are released. Or am I misinformed?
Where are we going and why am I in this handbasket?
Well Alphas broke in to that clock speed only after they went with coper interconnects witch was not fully developed until well after the P!!! came out(P!!!'s start @450Mhz.) Yes back in the 486 days RISC prossers had a higher clock speed and deeper pipe, but since the PC50 CISC prossers have had higher deeper pipes Intel knew that clock speed was the only way to catch up with many of the RISC prossers.
By the way the best I86 branch perdiction unit ever had a 99.4% it was used on the AMD K6-2/3, but held back the clock speed so it was dumped for the standord ~99.1% Bevlive it or not that makes a big diffrence.
It's a sad commentary on the community when articles on slashdot equate CPU clock speed to performance. Anyone with a clue knows that this is not the case, CPUs are complex systems with a broad spectrum of performance characteristics. Just compare Pentium 4 performance to Pentium III, or to Athlon, it should be clear by now even to casual observers that clock speed is not the sole factor involved in determining processor performance. A 900 MHz part from Sun should be a very impressive chip.
With the same switch, world peace will come, britney spears will have some talent and just maybe, code redworm will stop being a "threat". Ok ok.. I went overboard witht he britney spears thing, sorry ;)
-
ping -f 255.255.255.255 # if only
Everyone seems to assume that a basic Sun workstation will set them back 5-10K. I just checked the website and you can get an entry level system for about a $1000(US). Sun is a bit server-centric though.
Not true at the moment. These simulations were run on RedHat 6.2 running on 1Ghz PIII's. EDA vendors are currently dumping all NT support like crazy and jumping to support Linux. Most chip suppliers are setting up massive (100's of PC's running Linux) simulation farms for the above reason. PC's are cheap. Linux is great and for EDA apps it screams compared to SUN solutions. SUN makes nice machines and great OS's. Just way out of line in a cost/performance tradeoff.
"TV, a medium as it is neither rare nor well done." Ernie Kovacs
I don't know what APPLICATIONS you people are running but it seems all you are comparing are SPEC marks. We run Cadence Verilog and Synopsys synthesis on our Dell PIII 1Ghz machines and they blow the frigging doors of all of our SPARC's. For IC design applications at least 64Bit vs 32 Bit is a non issues as most of the apps are not compiled for 64bits. Performance is mostly just CPU clock speed so a 1Ghz PIII runs at just over 2x a Sun Ultra 450. Given that there is about a 3-4x difference in cost and about a 2x performance bonus you get about a 6x-8x cost/performance benifit. I'll take Intel machines all day long. Sun better wake up or they are going to lose the CAD market. Course, I don't think they really care about that one anyway.
"TV, a medium as it is neither rare nor well done." Ernie Kovacs
I would say that if the machine-level hacker knows the CPU architecture and the application down to the specific instructions, it is very often that the hacker can beat the compiler. Take a look at this optimization of RC4 and tell me if you change your mind.
there's next to no advantage to using vliw in modern cpus. all it does is make instruction to exec unit assignment simpler. everything that a vliw cpu gains from a good compiler a risc cpu will also gain. it's just that vliw is extremely ineffiencent without a very good compiler.
running what Mr. Enduser? sure it will be faster with one damn user on it
And comparision with _just_ a P4 isn't going to be much of a believable benchmark.
I'd like to know how soon they're gonna release a cheapo Blade 100 using this processor. I was considering picking up one of the 500 mhz models, but now I'm definitely gonna hold out for the next couple of revisions of the machine.
A fairly comprehensive overview: http://www3.sk.sympatico.ca/jbayko/cpu.html#AppA
I don't know where your hearing that but EDA tools are virtually all produced for Solaris, and HP-UX. NT is actually what EDA companies tried supporting but found the low demand. Nobody actually uses NT for real IC designs. As for Linux, EDA companies are porting as many products as possible to it. Applications run twice as fast as on the fastest UII and they're many times cheaper. In fact I was in a meeting a few months ago for the group in my company (one of the major EDA company) that determines which platforms to support and they were absolutely impressed with the performance of x86 on Linux. There are a few known kernel bugs to work out in Linux for these long jobs but overall people are very happy with it.
Ok I'll bite...
UltraLinux
Although they're supposed to compete with Intel's Itanium chips, they only run at 900MHz ... for now
Yes, as many others have pointed out, 900MHz high-end RISC chips should run comparably well to CISC chips with a much higher clock-speed.
Note that, in one of Intel's benchmark of their Itanium chips, an Itanium 800MHz was tested against an UltraSPARC-II system running at "only" 450MHz.
Of course, the results of this benchmark "show" that Itanium systems gave 12x more throughput than the UltraSPARC ones. Still, though, the benchmark was done with those chips because the chips are thought to be equal in processing power (taking into account the CISC speed to RISC speed translation).
mrcow
Ok from what I understand about how processors work (very little) generally the more transistors the faster the chip is (I think). So what would happen if they doubled the size of the chip, and thus doubled the amount of the transistors, rather than attempting to fit more in a smaller space, would the new chip run faster? I know this isn't an extremely desirable solution, I'm just curious.
"
Although they're supposed to compete with Intel's Itanium chips, they only run at 900MHz ... for now.
Just becaue it runs "only" at 900mhz doesn't mean anything compared to an Itanium running at a higher clock speed. There are many more factures like pipelines, cache, and over all archetecture. A 900mhz sparc could beat an Itanium at a higher clock speed just like Athlons and PIIIs can beat P4s in certain benchmarks while running at lower clockspeeds. (not saying it will or will not, but you can't discount one processor based only on clock.)
--------
It's OK to be social, just don't tell anyone about it.
They make all of Sun's UltraSparc chips, and also manufacture other, more esoteric things - like dual core chips (DSP and ARM, known as OMAP).
All in all, TI is much, much more than calculators.
Ok, enough Intel/X86 bashing. I will agree that the USIII is better for SMP, but everyone seems to be ignoring the cost. Let's compare two machines that cost around $3k. A 1.4 Athlon right now cost around $250. What does this new Sparc cost? A hell of a lot more than $250. I can build an X86 server with IDE RAID for under 3k that will perform well for most server applications. Can you build a Sun box for that?
Hmmm a Sparc 250 right now is going for 5k... It has a Ultra2 (not an Ultra3) at 300mhz.... I am willing to bet that an AMD 1400 will beat the snot out of that machine for half the price. What if you wanted to upgrade to another processor in that Sun box? Well that will cost you another $1500. For that price you can almost buy another complete X86 box.
I am a huge Sun fan, but for most small servers, you can't beat X86 with Linux.
Steve Michael
Windows is bloated because MS piles feature onto feature. The features don't work together, so there's a lot of implementation redundancy. If something goes wrong, a kludgy fix is added, making things worse. Everything gets totally redesigned every 6 months, so there's a lot of backward-compatibility support -- more implementation redundancy.
I thought Itanium ran at 800mhz max....
-Colin
Correct me if I'm wrong, but isn't the Itanium a CISC-based processor while the sparc is a RISC-based processor. Even at 900MHz, it should outperform most CISC-based processors. Or should I stop posting at 8AM?
Yes, but are they compatible with high speed fish?
Bullshit. SunOS is 5+ years old. It is like saying M$ is dying because there in no NewsNet posts from DOS boxes. Solaris is server OS, there are few workstations but millions and millions of servers in corporate environment.
At least with the G4 vs P4 debate, people can bring out benchmarks. I don't think anyone's put an Itanium up against this Sun bad boy yet, so all this article accomplishses is start a minor war, and no one will move an inch unless they have to, since there is no real proof available to make them change their minds.
It's just like religion.
Actually, it's using UltraSparc IIe chips. It's basically a USIIi with the bus interface from a III (based on conversation with a Sun engineer). It was done as an experiment to get the front side bus right, probably for the IIIi, and they figured out that it was better than the IIi, used it in the Blade 100, the X1, and IIRC the T1.
-30-
http://news.cnet.com/news/0-1003-200-6728520.html
I agree, but it's infuriating to see Intel's chips and architecture trashed in comparison to Sun's based mostly on problems with the Windows OS. Yes, Sun's stuff is far superior, but Intel's, while not elegant, is pretty damned fast.
I would not think of running any of their newer bloaty code on a 486.
That's okay because it's not an option. Their setup for ME tests to verify that you have AT LEAST a 150mhz Pentium. Anything less and it refuses to install (there is a secret switch to skip that test).
Hmm... I don't work on the Ultra 10's on my desk because they're too slow (and only 512 megs RAM, not enough). I personally do my development and testing on Ultra 80s, an E420, a couple E450s and 4 E6500's. I don't know what applications you're running but groups in my company have been benchmarking systems running Linux with PIIIs (~1GHZ) and they are blowing away the UltraSparcs. 1.5x to 2x faster in the type of applications we produce. I'm not sure why you're comparing the UltraSparc II to a Pentium anyway. Did you mean to refer to a PIII? The UltraSparc II and PII and PIII are almost the same age. We also have tests systems running with the UltraSparc III at 750MHz but I haven't played with them enough to get an idea of their speed.
uh, more bits doesn't mean a faster cpu. in fact, it means pointers are twice as long, which means they take up twice the cache and twice the memory bandwidth. the fact that most 64bit cpus are faster than more 32bit cpus has absolutely nothing to do with them being 64bit.
I much rather have a 900Mhz 64-bit processor that works over Intel's abortion. Plus the fact that the UltraSparcIII is superscalar to 1024, leaves Intel eating Sun's bits.
Part of Sun's success is how well they address the bus/throughput issue, as opposed to 'other' computer architectures. And that's why JUST comparing MHz is like comparing apples and oranges.
Or perhaps a better anology is comparing a Formula 1 Racing car stuck in down-town NYC Traffic, versus a 6 cylinder Honda Accord on flat, wide-open highway in Montanta, during the daytime when the weather is perfect.
healyourchurchwebsite.com - WWJB?
can you say 8Megs of cache? as intel chugs along with 512k, maybe a Meg...need anymore be said?
Unfortunately, this is what a lot of managers will see: "Here, this Intel system runs at 1.8 GHz! Why are we buying this slow system which runs at half the speed?" People who understand systems see this is patently crap, but it's what the non-techies see. Even IBM systems are only just getting up to 750MHz, but they still kick a lot of Intel systems into touch as far as performance goes.
Here's a tip: If your comparing CPUs, benchmarks are basically all the information you need to know. If you're comparing systems (which means real world loads) then that is where benchmarks fail because not one of them can reproduce your specific real world load. In addition, it's not the UltraSparc CPUs that you want on your "mission-critical" applications but it's the huge cache, and the rest of the hardware that surrounds the (by my personal experience and estimation inferior) CPU.
IBM's older Power2 processor is probably the ultimate example of "mhz doesn't count". The version I've got in a machine runs at 66MHz, but smokes a 200MHz 21064 Alpha. (then again, it was 6-way superscalar and loaded 8 words from memory per clock) Heck, in my own usage, it "feels" like a low-end P2 (and probably is just as good or better on FLOPS).
And this thing was made in 1992!
Where the real advantages come in is with things like memory architectures (eg, memory interleaving) and bus speeds (where the system bandwidth is more than an x86 solution) which is relevant in databases. Added to that, you can scale these up much more (the E6800 can have 24 900MHz CPU's, for instance; Fujitsu have recently released a 128 CPU system based on their USII clone at 500+MHz).
If you want a measure of raw CPU performance, check www.spec.org; currently, the fastest single CPU systems are Intel P4's (although some alphas come damn close). The Sun 280R doesn't come close to that, although it is faster than its clock speed would suggest...
It blows any of the pentium-based machines we have here out of the water.
-fialar
wow, someone badmouths a 900 mhz chip from sun being competition for the Itaniums (which, i read somewhere, were only running at 800 at this point...), and all of a sudden the "it's the pipeline, stupid" and "there's more than just megahertz" people come out of the woodwork.
i'll have to remember this when the next G4 vs Pentium4 debate comes up...
- Entertaining Bits from the Ancient Kernel Tree
"Seems?" Exactly what type of scientific tests did you perform to find such conclusive evidence as "seems?" Let me guess -- your Windows box, which is optimised for the desktop, "seems" faster than CDE on the Sun box, which runs over an OS optimised for REAL computing. You little dipshits refuse to learn...
----
Please, I are begging you! To save Dmitry from teh jail!
It's not how fast the processor runs that is the most important performance factor, it's how much it does per cycle.
Zero Kelvin
neux.org
Could you even tell anyone what the fuck the difference is? I think not. Why don't you go back to writing Perl for web pages.
I work with Ultra 10s, 60, and 80s daily. From the normal work, UltraSPARC chips do things about twice the speed of a similarly 'clocked' Pentium chip.
UltraSPARC 450s do things about the same time as Pentium 900s, etc.
These should be screamers. Don't be fooled by the number attached to the chip.
DanH
Cav Pilot's Reference Page
UNIX - Not just for Vestal Virgins anymore
one thing i have noticed about Sun hardware as compared to Intel hardware is that the Sun stuff will take quite a load without crumbling under the pressure. my experience has been that whether you've got 1 user or 100 users on the system, the Sun stuff pretty much runs the same. with Intel hardware, your results will vary, especially among the different OS's, but overall, i don't think Intel hardware handles a heavy load as well as Sun equipment.
No really.... look at the RISC vs CISC threads. Also, you need to compare the rest of the system design. I'll put a 300Mhz Ultra SPARC II in a old e250 up against a 1.4 GHz PC as an email server any day... the e250 will blow the pants off of the PC.
Back when AMD processors were running at higher Mhz than Intel processors, everyone in /. seemed to be singing a very different tune... the opposite of the current song, in fact. Its' amazing how fast everyone has gotten educated to the reality that Mhz is a foolish metric...
Itanium only runs at 800Mhz so what part of this chip running at 900 Mhz is slow?
Jon
How in a civilised society can we sit back and let this apocolypse happen? I say its time to end this now. Boycott processors. Save the instructions
Why would anyone with half a brain buy expen$ive Sun hardware when you could get beautiful 4 or 8 way Intel-based Compaq or Dell boxes for half the cost and over twice the speed? It's ironic that Sun's Java push will ultimately lead companies to buy inexpensive non-Sun hardware to run the language they invented. When that day comes and Sun's hardware sales fall off a cliff - do you think J2EE will remain free? Don't bet on it.
'No! We're under budget cuts right now!'
"But c'mon! I really want one! They're s00pur c00l!"
'No! You'll stick to your microSPARC and like it!'
Kinda makes me wish I didn't transfer from my last job.. THEY would have gotten me one...
Moral of the story: These things may be cool, but they're so expensive only your boss could buy one for you.
Never hit your grandmother with a shovel, for it leaves a bad impression on her mind...
Attempting to measure how fast a computer can go by its CPU's clock speed is tantamount to measuring how fast a car can go by its engine's horsepower. There are many more factors at play here.
Let's start with the whole RISC vs. CISC thing. Everyone knows that RISC is more efficient; the only thing that has kept CISC alive this long is backwards compatibility with the Wintel juggernaut. You develop a lean, efficient instruction set, then you write compiler back ends that take advantage of it.
Also keep in mind that Sun's motherboard designs are true performers. The path between the CPU, memory, and bus are designed to move data around in ways that just aren't possible with Intel.
Did you know that SPARC is more or less an "open" CPU design? It was designed to be a multi-vendor instruction set, one that would be 'common' without having one vendor calling all the shots. Read www.sparc.org for more details.
Tired of FB/Google censorship? Visit UNCENSORED!
At least for ssl transactons. Check www.coradiant.com on www.coradiant.com
One of the reason that UltraSparc kicked the crap out of Intel P series chips is that that intel chips are 32bit and the UltraSparc platform is 64bit. A similarly clocked Itanium and US3 is going to compete in the same performance range because they will both be 64bit chips. The old rule of thumb that a Sun chip is twice fast a similarly clocked Intel chip is history.
- Dustin -
And it doesn't enable the Internet either.
I'm a little surprised a technical web site would fall for the pure marketing hype. Next we're going to have an article complaining that the Ultra SPARC IIIs run only at 900 MHz and can't play the new space cadet game. That is a fun game it wasnt free but it was worth the money.
Insert simplistic political, ideological, or personal proselytization here.
For all practical purposes, Slashdot is dead.
People not using Solaris is killing Slashdot? Damn those IRIX users!
Solaris went out of business
Did it instantly delete itself from everybodies harddrives when it went out of business?
Sun...sell another troubled OS.
Since when did Sun replace Solaris with BeOS?
A recent article put SunOS at about 80 percent of the Solaris market
My head hurts. See I would put SunOS at about 100% of the SunOS market, and Solaris at about 100% of the Solaris market. Although we all know now that Solaris has gone out of business, and Sun now sell BeOS. If you're interested, i'd also put BeOS at about 100% of the BeOS market.
Now slightly offtopic, who the hell is Kreskin anyway?
BTW, I heard three months ago they had prototypes running at 1.5GHz. I guess there still working out the kinks.
Now these are the kinds of articles and comments that I enjoy. It's on the techy side and not fluff pop culture material or the latest craze or fad in overclocking. It also seems to bring out the more intelligent people in the group for real comment. Not just some zealots praising an OS over another or just personal opinions. Keep these kinds of articles up!
There was something linked from solaris central about how the Ultra III was for massively multi-processor machines, not workstations. The Ultra IIIi will be the workstation processor clocking in(I think) at well over 1GHz. I'd provide the link, but I can't remember it.
-blah
The following sentence is TRUE. The previous sentence is FALSE.
What makes Spec so much better than all benchmarks(no sarcasm here, i'm really interested) and why do so many people point to Spec when it comes to checking processing power? Are they anymore untainted than any other benchmark? What other benchmarks are out there that maybe have a different set of priorities than spec? If an Athlon or P4 doesn't win a benchmark, is it invalidated? Isn't it possible that a processor can completely suck at a certain Specbench and actually be fairly good overall? In other words, what kind of correlation does a spec score have on the performance of my computer or a set of computer systems in everyday use? I say throw the benchmarks out. For most people using computers, it is the operator that is causing the greatest bottleneck. The best advances in computing have been those that help the operator become more efficient.
When are people going to get the clue that clock speeds don't really matter unless your comparing clock speeds of a CPU from the same family? I mean, a P4 1.7GHz is way slower than the PowerPC G4-867 or whatever they are up to. Here is a link to a little tidbit that John Rubinstein from Apple Computer gave at MacWorld NY the MHz Myth. http://www.apple.com/g4/myth/
A fully loaded RS/6000 B50 can be had for half the cost of the nearest Sun box, an E220R. You are right in that IBM cannot compete with Sun's 1U offereings, such as the Netra X1 ($995).
I've worked with Solaris and AIX, and while the RS6Ks are much better technologically, AIX is of no match to Solaris. Once you get used to a SYSV system, such as Solaris, you can move around to HPUX, SCO (god forbid) and Linux with ease. As for AIX, everything is so outlandish and archaic. Don't even get me started on the ODM, or the idea to use "stanzas" in config files. Let's not forget the mentality to create a command for every piddly function, some of which could be accomplished by editing a file. mknfsexp is a prime example.
If only Solaris 2.51 PPC had been successful, we could have had the best of both worlds. Sun doing the software, IBM doing the hardware.
The UltraSPARCIII chips running at "only" 900mhz is still much faster than a Pentium class chip running at equivalent speeds. This is completely different architecture than x86.
Alternatively, read this.
Just look at the requirements to run the various Windows OSs. When Windows 95 came out, the bare minimum to run it was a 386DX at 33mhz, 4MB RAM, and a 100MB hard drive. Windows ME requires, at a minimum, a 150mhz Pentium, 32MB of RAM, and 480MB of hard drive space. The RAM requirements have quadrupled, the hard drive space has gone up by a factor of five, and CPU power has gone up by somewhere around a factor of 10. (I know that there is some disagreement about what the actual minimums are, but I believe these to be in the ballpark and they illustrate my point.)
So, if you want to find out what the CPU is capable of, dump the OS, write an application that taxes the CPU, and run it on each. (No, you do not need an OS to run a program.) Until you do that, you're just tossing around meaningless numbers.
Imagine a Beowolf Cluster of THESE!!!
According to my experience, my 100Mhz whizbang chip with a 512 bit bus, can roll up you P4 1.4ghz and smoke it!
I'm seeing alot of SPEC this and SPEC that but unless I'm mistaken most people buying these machines are way more interested in TPC. So until I know how Oracle behaves on a latest-generation Sparc box running Solaris versus a comparable HP box running HP-UX or [insert architecture and OS here], I'm not placing any bets. Besides, as long as price/performance stays reasonably close, Sun users will stick with Sun and HP users will stick with HP and so on. There tends to be too much of an investment (in terms of in-house knowledge, existing software, vendor relationships, etc.) to compel anyone to switch over a relatively minor difference.
Please donate your spare CPU cycles to help fight cancer and other diseases
The fastest Itanium chip right now is 800 Mhz. A lot of people have been railing (correctly) against Mhz vs. Mhz comparisons, but in this case, it isn't even necessary. Itanium is not the same as a P2/P3/P4. It is a totally different architecture, and its clock speed currently tops out at 800 Mhz, not 1.7 Ghz like the P4.
If you are going to attempt to troll, at least make the post somewhat comprehensible. Most of what you said had no flow, contained too vague references, incorrect logic, and complete
mis-information.
On the other hand, I do see Sun as a troubled
platform, as I have never been a fan of SunOS/Solaris, and the hardware costs are beyond any reasonable price for the not so spectacular performance. At my last job, they put a brand new, $8,000 dollar Sun workstation on my desk, and it was so damn slow compared to much cheaper PC models, they really gotta get their act together, paying through the nose for a crappy system is unreasonable.
I do know that Solaris on SPARC is pretty robust, and I can jerk it around without it complaining. A Solaris/SPARC system is also very resilient when hardware fails (if the system is set up for it). The reason for this is that Solaris, SPARC processors, and the rest of the computer are in a sort of symbiosis--they are designed to work together very well (and they do). That's why I prefer my 9-year-old SPARCStation over my more modern PC; it is just a better architecture to work with.
I also know that this is generally not the case for Pentium-based systems. Many Pentium-based servers look and behave as if they were engineered by monkeys--especially when they are forced to suffer under Windows. Will this improve with the Itanium?
Healthcare article at Kuro5hin
well I have to say that the HHGTG was on the googlebox last night very shocking
and amuseing
the improberbility drive had me in fits of laughter proberly
ah well if only that women had not forgotton how everyone could be happy and live in peace
damn her
regards
john jones
What's lost in all the AMD vs SPARC vs P4 debate is what horrible SPEC INT performance the Itanium has:
The 800Mhz Itanium has the same Spec Int 2000 performance as a 800Mhz Pentium III!
And that's with 128 registers and all the "workstation" bus piplines fast memory and everything else vs. the Intel reference motherboard for the Pentium 3.
Yes, the Itanium has good SpecFP performance - but I bet a couple Ghz Alphas would have spanked the Itanium workstations HARD using old tech chips that have been available for quite a while.
All we can hope is that Intel doesn't buy the Alpha and kill it off.... oops... too late.
=tkk
Bill Gates - Creationist?!?
at least SPARC encourages chipmakers to implement their own SPARC-clones, ever heard of Intel being happy with AMD x86 clones?
Wow.. My Mhz are faster than your Mhz... Come on guys... I bet you run WinBlows too... Are you even old enough to know what a 64 bit machine means?
Comment removed based on user account deletion
Wow, imagine a Beowulf cluster of these...
It was hard for me to run a reasonably configured Win95, release 1, on 66Mhz 486 with 500MB hard disk and 16MB ram. Another 500MB hard disk helped. Later MS ware, while having needed fixes like the ability to see much larger hard disks, use MSIE which wants a 500MB hard disk foot print. I would not think of running any of their newer bloaty code on a 486.
Friends don't help friends install M$ junk.
There is a lot of confusion as to what RISC and CISC are exactly. However, there are some features that all RISC chps have:
1. Fixed length instructions
2. Load/store architecture (only load and store instructions access memory)
Anything else has already been adopted in CISC chips, such as pipelining, caches, etc. As far as I know, these two things are still the main differentiators (all RISC chips share these two things).
Apples vs. Oranges: if that is true then you cannot compare PIII to P4 to AMD (all different architectures), but such comparisons are done daily on various PC hardware sites. You can compare the performance of architecturally different machines, but not solely on the CPU clock speed (which by itself isn't very informative).
Also, saying things like: "the cpu is slower but is has a better memory architecture/bus" ignores the fact the memory architecture is largely determined by the CPU's capabilities. I doubt that many x86 processors can support 8MB of 2nd level cache which alphas can. Does this mean the alpha has an advantage? Of course it does, just as it would if it had more execution units, higher clock speed, etc. Interestingly, it is now CISC (x86) that needs the high clock speed to maintain performance (P4).
Finally, it is interesting to note that spec's benchmarks indicate that the P4 and even P3 are faster than the athlon, but many PC hardware sites claim otherwise.
I run the same applications on Ultra 60 (Solaris) and on 1GHz Pentium III (Linux). These are EDA tools from Synopsys and Cadence, integer only. Programs with small main loop footprint (work from L1 or L2 cache) run faster on Pentium, proportional to the clock speed, i.e. two times faster. However, when there is a lot of random accesses all over the main RAM, Sun hardware runs slightly faster. Looks like it has better RAM subsystem.
This is orders of magnitude more funny than the parent...
The UltraIIIs come with nice new toys, too - they're available in the E3800, E4800 and E6800, which can all run a cluster-in-a-box, amongst other things.
Also, many people are still using Solaris 2.6 because they like it and see no need to change. Because the UltraIIIs require Solaris 8, people get to see the new toys in that, too (IPMP - redundant network connections, and other such goodies)
discl: I work with Sun, though not for them.
Author, Shell Scripting : Expert Re
My 1993-era POWERserver 590 (66MHz Pwr2) really blew me away when I found these results (MB/s) in the STREAM benchmarks. And yes, I did reverify them. Though you only get these numbers with at least "-O5" to the compiler, so it could be doing something funny. However, without super-optimization, it does get at least half these numbers, which is still pretty impressive.
Machine ncpus COPY SCALE ADD TRIAD
RS/6000 590 1 600.0 533.3 654.5 654.5
Register windows are a double edged sword. When we run out of windows, we overflow which means pushing 16 registers onto the stack, which at the best of times implies at the very least 16 mem accesses (not to mention stack checking etc).
With only 8 register windows, there is not an OS/app in existance that can perform even the simplest sys call without causing an overflow (or ten).
Don't even THINK about context switching...
internal cache stores the undecoded instructions, so you still have to decode them each time. This only matters if your decoder can't fill the pipelines on the chip.
Incidentally, the P4 stores decoded instructions in a cache, but it can only issue 3 of those instructions each cycle, so you're guaranteed to have an underutilized chip
Reboot macht Frei.
18 comments in and somebody caught that. Bugs the hell out of me when I see that in a chip review. Maybe someone will invent a 3ghz chip next week that uses 4 t states for every internal cycle (read 750mhz). I'll bet the press would hype it up and the investors would pour in the money, and we'd all be rich before we had to get it out of the fab... hold that thought
As you implied, SMP performance is extremely important to people who buy Sun.
In this case, you wouldn't care much how an individual processor performed; you are most concerned with the performace of, say, a 32-way system and it's ability to quickly shuttle data between processors, memory, and disk.
Our beloved Athlon only scales to 2-way, and it's SMP architecture is now being entirely redesigned with the NUMA hypertransport.
Sun probably suffers in raw MHz and SPEC scores because they put so much effort into the SMP aspects.
And, of course, Sun outsells some (arguably) better technology (Power, Alpha) because they are much more open and their service organization is superior.
Correct me if I'm wrong, but isn't this what an internal instruction cache is?
Your Servant, B. Baggins
If you want pure RISC Alpha is the nearest thing to it. In fact one of the RISC pioneers (can't remember the guy's name ATM) said a few years back "RISC processors have got so complicated that the original design goals have been all but lost".
Once you start to add features like out-of-order execution, branch prediction, register windowing, speculative fetching, etc, etc your chips start to get complicated and won't scale as well hence is the opposite of RISC philosophy of keeping the design simple (a by-product of reducing the IS) for better performance.
Compiling on our Itanium (667Mhz) machine is like make world on a 386.
IIRC, 1992 was when work began for the first iteration of WinCE and I believe MIPS was the only CPU supported initially, with MS adding the other CPUs as they went along...
Perl - $Just @when->$you ${thought} s/yn/tax/ &couldn\'t %get $worse;
But what good are these chips? You can't run Windows NT on them!!
how come someone like Yahoo doesn't go around bragging about the 450 MHz ultraspace chip they have running the website.... because it doesn't matter
That Yahoo uses FreeBSD/Intel is probably another reason.
cpeterso
I also noticed (over the past few years) that Sun has been leaning on the UltraII since like 1999.
:)
Back then they were way ahead of the game, since Intel/AMD were doing about a quarter their current clock speed, while Sun was doing about half.
I wonder what happened