Slashdot Mirror
Sun's New Workstations and Graphics Cards
An anonymous reader "Sun Microsystems has released the Sun Blade 2000 workstation, along with a new graphics accelerator, the XVR-1000. This could very well give SGI's lineup a run for its money in the CAD and Visualization fields, although its fillrate and 38-bit colour may make it less desirable for animation. Make sure to check out Ace's article. " (page down
a couple times to read it)
How does that break down storage wise?
http://news.com.com/2100-1001-860701.html
http://www.sun.com/2002-0314/feature/
The system ships with a 73gb fibrechannel harddisc, 900 or 1.05 UltraSparcIII (dual capable), and a gig of ram. nice box. It sets a world record in workstation performance (halfway down the press-release).
Not flame bait, but a legitimate question. What would someone be using a $34,000 workstation for? Even a $9,000 one?
They can't possibly be selling THAT many of them.
Anyone here using them? What for? Is a PC really not that powerful?
Direct link to the post as a stand-alone page.
I have Sunblade 100, from which I write this comment. From my experience, this machine is by a factor 4 (yes, four) time slower than a new Athlon XP 1.9... And it costs much much more.
If it wasn't for endianness compatibility with existing binary data, I wouldn't be using it.
and at only $11K its a steal.
Or rather, thats the only way I'm getting one, theft.
--"Karma is justice without the satisfaction"
Sun never needed to answer to Intel's 64bit cpus. Sun corners a market that Intel has not even begun to penetrate yet.
Just the fact that Sun and Alpha have been doing 64bit years illustrates that fact.
Also there is a little bit of a misconception here. They perform drastically different because of the SMP bus architecture and just the fact that it's CISC vs RISC etc.
Sun couldn't beat Digital? I'm confused, I thought digital was getting beat, got bought out and their new parent company killed them. Intel has just started to enter the server market it a serious way, so your suggestion that Sun has already lost to them seems unfounded at best. Microsoft has a tiny share of the high end server market which Sun prefers, so I think the jury is still out on that as well. Is your whole arguement based on the fact that Sun isn't dominating the home computer market?
Pushin' 'n dealin', shovin' 'n stealin'
What do you mean 38 bit color makes it less desirable for animation?
That is just wrong. This has 10 bits per component RGB. Typically that's more than enough. In addition animation apps like Maya tend to be geometry and state limited not fill limited.
Ofcourse the tag 'animation' is a bit to vague to mean anything in the first place.
Well done Sun, this should cause SGI some pain, but I'd say more because it gives the impression that Sun is doing something interesting where SGI hasn't done anything genuinely interesting in a LONG time.
I dont think that intel ever really had a question for Sun to answer. At 800MHz, Intel's 64it chip is slow in the all important MHz rating (sun has had 900's out for a while now) and still has a few years of compiler design ahead of it before it makes any sense. And this is Sun's 3+rd generation of 64 bit chips, vs Intel's 1st.
As for HP, they helped intel build their 64 bit chip, so the PA-RISC is more or less dead.
I'd say this is the misconception. The advantages of RISC over CISC for an equivalent clock speed CPU actually vary significantly based on the TYPE of workload. A good example: a while back a customer was complaining that compiles went twice as fast on their HP PC platform (1GHz CPU) than they did on their Sun platform (450MHz CPU). Compiles are almost entirely CPU bound. Found numbers point out that the SPEC ratings for the 1GHz CPU were about twice those of the 450MHz. What a surprise.
The thing is, the machine with the 450MHz CPU had 4 CPUs. If they had invested some effort in configuring a parallel make, the 450MHz machine with 4 CPU's would have approached being able to half the compile time of the single threaded make on the PC.
7 November 2006: The day Americans realized corruption and incompetence weren't addressing 11 September 2001
>> although its fillrate and 38-bit colour may
.plan once.
.plan about this...
>> make it less desirable for animation
I'm not sure what that's supposed to mean. The fillrate is just fine for a workstation, games generally are the only programs that need a high fillrate, memory bandwidth and size, and of course T&L are *much* more important. The 38 bit internal color is excellent, nicely comparing to SGI ( http://www.sgi.com/workstations/comparison.html ), and unmatched by 3dlabs. The bit-depth of the graphics card has nothing to do with the color rendering accuracy, which is usually 48 or 64 bits for high end stuff. Games really need high bit depth precision for multitexturing, which multiplies color errors. I think Carmack mentioned this in a
Nvidia will probably have 64 bit color in NV30, and 3dfx's rampage was supposed to have 52 bit color ( http://www.digit-life.com/articles/3dfxtribute/ ) Games start needing high bit depths when you have massive multi-texturing, which tends to multiply errors. I think Carmack had a
Where I work, we need computing power for 2 things:
1. Running builds
2. Simulating embedded processors (ARM, mcore) for testing our product.
We have a mix of Sun workstations and x86 linux boxen. We just got one of the new-ish SunBlade1000 for trial (single 900Mhz processor, 1GB RAM).
While the Sunblade kills the competition (1Ghz Pentium4 w/linux) in build times, it's actually slightly slower with the simulations (which were, ironically, developed natively for SUN architecture!)
So, before you think about getting one of these puppies for your own pad, you better find some published benchmarks specific to your needs. There's no magic bullet.
Free unix account: freeshell.org
D'ohhh! It's HERE!
The low end really has eaten the high end in graphics hardware. Five years ago, the $1000 boards outperformed the $100 boards by an order of magnitude or more, because the high-end boards had hardware Z-buffers, geometry hardware (the 4x4 matrix multiplier), and hardware texture and lighting support. Today, low-end 3D boards have all that; the high-end boards just have a bit more of everything.
The cost probably reflects about $400 in parts, and millions in engineering cost divided by the few hundred of these boards Sun will sell. That's a losing business proposition.
Sun also announced a 24" high-resolution flat-panel monitor. Any info on that?
Very interresting If they Pull it out.
--- Metamoderating abusive downgraders since my 300th post.
Doh. That would be everythingunix.ORG. Or go straight to the article here.
Does 10 or 12 bits really make a difference over 8 bits? Of course it does. Most film work these days is rendered in either 12 bits, 10 bits logarithmic, or 16 bits. Think about it: in a dark movie theatre room, 256 levels of grey (for instance) is not a lot. And if that doesn't convince you, think about image manipulation: after a few multiplications and compositions, you'll end up with very little color resolution with 8 bits. And yes, these things are often done in hardware in the color buffer (eg flame).
Sun FastEthernet 10/100Base w/ MII -- $695
10X DVD-ROM SCSI based-- $400
73GB 10,000RPM FC -- $4,100
If you look at the SunBlade 100's options its even scarier.
16X DVD-ROM - IDE based -- $295
20GB 7200RPM EIDE -- $300
Now I absolutely love Solaris and Sun Hardware, in fact, I'm using an Ultra-10 now. I just think these prices are a little out of hand. Especially when you take the time, I'm not about to take, to look at the costs of the same items from the actual manufacturers of the products. The Quantum manufactered, Sun branded, DLT drive I installed yesterday, cost $1000 more for the purple die job and Sun logo.
"The words of the prophets are written on the Slashdot walls."
I personally feel there is a more interesting article that is linked to: http://www.sun.com/executives/realitycheck/headsup 020314.html
details MHz-vs-Speed differences. While not the most interesting for the well-informed, it's great for those who know that MHz doesn't necessarily = speed.
-- Sometimes you have to turn the lights off in order to see.
Hey, don't call my beautiful 100 a toy! :) Seriously, for me it's perfect, especially with a flat panel. And once my SunPCI card comes in in a few weeks, I can finally reduce my desktop to one keyboard and monitor and run in both unix and windows. Even though both are around 700mhz, nothing I do (run emacs, outlook, listening to mp3s, running a web browser, etc..) will even come close to taxing the CPUs.
_______
2B1ASK1
You really need to work on your understanding of "high end"; Sun's high-end is boxes like the E10K and E15K -- and it's an area where Intel has no leverage. An E15K can support multiple hardware domains, up to 106 US3 900MHz CPUs, and over a half *terabyte* of RAM.
You find me an Intel machine with those specs. Oh, and it must be fully managable from a remote site down to the hardware level; you have to be able to turn CPUs on and off, power the machine up and down, re-assign drive IDs, and such -- remotely.
The eight-way xSeries competes more with Sun's low-end server hardware, which is comprable in price; I can't really give an exact figure without knowing what this server is for.
--
I Hit the Karma Cap, and All I Got Was This Lousy
on this one. How the heck am I spose to find a bunch of "hardly used" sun sunblades on the cheap for personal use? In this new age of fiscal responsibility and limited cash, there is no way I can convince managment I need one of these as a MP3^H^H^Hsendmail server....
+++ UGUCAUCGUAUUUCU
Which "equivalent box"?
An 8-Way Sunfire with 32G RAM and 400G SCSI storage lists for $120K.
A Pirate and a Puritan look the same on a balance sheet.
Ace is wrong about one thing:
Currently, the XVR-1000 targets primarily the engineering and CAD markets, as opposed to 3D animation, given the rather limited fillrate. However, Sun intends to use the MAJC-5200 to scale the performance of its graphics solutions to higher levels in the future (as seen in this older roadmap), so we may yet see a solution attacking the 3D animation market at some point in the future.
The MAJC-5200 will improve geometry performance (number of triangles, floating point math required), not fillrate (number of pixels/texels shaded, integer math).
Animation requires better fillrate, and more MAJC-5200s won't provide that. MAJC-5200 *will* provide Sun with stronger geometry performance (FLOPS, remember?), which is just what Sun's core engineering and CAD markets most want. Lots of small triangles to accurately show the precise shape of things of digitally-created parts. Nothing about MAJC-5200 will strengthen Sun's penetration into new SGI markets per se. That'd be dependent on some other, presumably fill-rate enhancing, technology.
--LP
With some highly specialized systems they might've managed to up that a bit, but your average system is limited to 4 gb.
-
We have sun blade 100's at work, and they make great workstations. But being a rebel I wanted to put Linux on it. The only Linux distro with the best support was SUSE 7.3. Suse is a great distro, but they can only do so much without help from SUN.
Some major problems with linux on sunblades.
1. DMA doesnt work correctly.
2. GFX card drivers, only the basic onboard card is supported, dont get the high end elite cards.
3. Sound support is a hit or miss, sometimes it detects and loads the modules, havnt figured this out.
For a 1000 bux box, usb and firewire, dvd, takes PC memory for a SB100. If linux was truely supported, they would sell ALOT more.
Where do they claim that? According to the SPECcpu website, a 1.05 GHz SPARC III Cu gets 537 base SPECint and 701 SPECfp, while a 2.2 GHz P4 easily beats it with 790 SPECint and 779 SPECfp.
Intel is way ahead in integer, and although the Sun catches up somewhat in FP, if you look at the individual results, it's entirely due to one massive spike on the art test. They recently figured out a (controversial) compiler trick that gave them nearly an order of magnitude increase on that one SPECfp test, and doubled their overall SPECfp score. Sun are known for their stability & scalability, but not their CPU speed.
Of course, if you have 106 of the things, that's different. But you'll be paying over US$4M for it, which isn't exactly workstation class anymore.
Why would anyone engrave "Elbereth"?
If you're willing to spend the money to get the speed, the nVidia Quadro4 900XGL is the current SPECviewperf record holder, supports two displays (2048x1536 each, better than the XVR's dual 1280x1024), and costs well under half the XVR-1000. It also supports stereo viewing and a programmable vertex & pixel pipeline.
True, its DACs are 24 bits instead of 30 bits (SGI workstations are still the go there, with 36 bit RGB DACs), but the NV30 may change that. It also does multisampled anti-aliasing (currently 9-tap 4-sample, though older drivers did offer a 16-sample mode too).
Why would anyone engrave "Elbereth"?
I understand exactly where the extra cost is coming from. I also believe that there should be a cost increase associated with that "Peace of Mind". However, that "Peace of Mind", does not come at the cost Sun is pushing.
CTO's are seeing their budgets slip through their fingers like water in their cupped hands. When it comes time to acquire a couple of new web servers, their eventually going to turn to Linux, or *shudder* Microsoft. Cost is now an issue more than ever.
"The words of the prophets are written on the Slashdot walls."
I want one of these to replace my Matrox G400 Dual Head MAX card. :~(
/too much/ of a hit from running in 32bit color VS running in 16bit color. Now days a lot of cards run better in 32bit color then they do in 16bit color. . . .)
Sad that it is not PC compatible, though I can guess as to why.
Sun should seriously think about getting into the PC hardware business for the high end proffesionals, there really is more potential to sell peripherals for the wide PC market then there is in trying to get everybody to switch over to their plateform. (how ever kick ass their machines may be.)
Ah, besides, a G400 MAX card that could do a bit more in the 3D arena from time to time would also be nice, hehe. I would seriously like to be able to run the occasional game at a resolution higher then 640x480@16bit color (well actualy I can run in 32bit color since the G400 was one of the first consumer cards to not take
Ah, and no the G500 is not what I am talking about. ^_^
Oh well, hopefuly the Kyro3 will be coming out Any Day Now(TM), though I do believe that it is a year or so behind its unofficaly leaked due date, LOL!
Need help treating your acne? Come here!
> There are people who do visualisation who care about this stuff. As for the color, does anyone know if you can actually see any difference there ? I mean - 24 bit color is 16M colors ?
Yes, 24 Bit color is 16M colors, but that is *inadequate* when you start talking about color gradients. 24-bit color has 3 color channels, each with 8-bit depth. That allows for 256 shades of *primary colors*, but the eye can detect millions of shades. A higher color bit depth has less banding issues.
3DRAM has been around 4-5 years or so. It is nice technology, but to answer your rhetorical question, adding logic operations to memory adds significantly to the expense of the RAM. (It's a non-commodity part made by Mitsubishi, and at the very least must be tested, and I think manufactured, in custom ways.)
It also reduces memory flexibility; you can't just take some of that huge texture memory you have and start using it as the frame+Z buffer of a dual-head display for example, unless the right amount of 3D RAM was spec'ed in the hardware design to begin with.
Also, at least in the early days, some blending modes were supported and others weren't.
Reducing Z buffer bandwidth is pretty nice though, don't get me wrong. But most of the industry has stuck with the volume economics of more conventional RAM types.
--LP
Looks like the two may be comparable. The Fuel costs about $11,000 for a R14K 600 model. I think that the Fuels v12 graphics may have the edge here, but for slightly lower end stuff, I can see companies going with Sun (We know they'll be around in 5 years) instead of SGI for some of their MCAD stuff.
What Sun gives you is a bit more performance per processor, or a bit more performance per multiprocessor box. But that is not usually a compelling argument, since big computations are usually distributed anyway, and it's still cheaper to build a 200 processor Beowulf cluster than to buy a 100 processor SPARC box. (The Beowulf probably also gives you better I/O and memory bandwidth overall.)
64bit processing is not compelling enough to cause a lot of people to switch. With cheap memory, that will change over the next couple of years, but then AMD and Intel will have mature 64bit offerings.
Sorry, but Sun has been steadily going downhill. They just don't have much of a market anymore.
Actually, our Linux Beowulf cluster is in our machine room, next to a (now unused and defunct) Sun Enterprise server.
DEC had at least competitive hardware, you're right. Maybe better. But my point was DEC was no a case where Sun developed superior hardware and was beat.
The "Non-Stop" line is interesting though. This is the old Tandem product line. Tandem specialized in systems that never went down -- even if some of the hardware was broken. Not that impressive nowadays, but Tandem dominated the field 20 years ago. After the '89 quake, Tandem got a service call from a bank whose mainframe server had been knocked over by the first shock -- but was still running. So please tell us, how do we bring it back upright without shutting it down?