IBM

Thinking of grabbing a POWER server - Page 2

TeamBlackFox wrote: >You have some impressively bad luck with some hardware there.
That I do, with Intel/AMD chips at least!


Try Cyrix!
TeamBlackFox wrote: Never had a SINGLE failure from a RISC based device of any kind.

Then you haven't run SGI kit very much. Fuels fail with depressing frequency. Octanes are a little better, Indies go through their power supplies, the O3xx series has power supplies that catch fire, and on and on.

Meanwhile, the Assistant's Netfinity trudges on down the road, has not had a single hardware failure in ... no date on the tag. But it's Made in China and at least eight years old.

While we're doing anecdotes, I have always had better luck with IBM than the rest. Now that Ginny is in charge that's probably gone with the wind but they did have better quality in the past. HP was good in the days of the RS-20 but I wouldn't touch their shit with triple-ply toilet paper now.

Then I was looking at the late model Sun Ultras and the Sun Blades...

As long as you run it entirely from the command line, Sun is nice. But every desktop they've had since NeWS is disgusting garbage. I'm pretty sure OpenStep won't run on a modern Sun, but if you get it to work, let me know !

Can you run Rhapsody on a G5 ? That might be decent.

You realize, the most practical thing would be Windows XP on a Z-Pro.
The time has come for someone to put his foot down ...
Kira wrote:
ClassicHasClass wrote: I despise the x86 ISA as well, but Intel has had a lot of money and time to invest in making it run well despite its warts, and in fairness to Intel they've tried to kill it at least three times (iAPX432, i860/960, Itanium) and the market wouldn't let them. So I can't really blame them anymore though I used to.


Offtopic historical note: 860 and 960 were completely distinct designs. 960 was a combination of a Berkeley RISC with some concepts from the iAPX 432 - notably, in the beginning, tagged memory; on the other hand, 860 was an odd proto-VLIW chip with no relation whatsoever to the 960 and an emphasis on HPC and graphics.


Editorial point conceded; I didn't mean to conflate those in my post.

hamei wrote: Can you run Rhapsody on a G5 ? That might be decent.


No. I did get it running on this Wallstreet G3 and, well, it's NeXTStep with Platinum but no app compatibility. Also, the compiler is a freaked-out gcc 2.7.2.1. I like it in theory, but when I use it, I end up spending all my time in the Blue Box.
smit happens.

:Fuel: bigred , 900MHz R16K, 4GB RAM, V12 DCD, 6.5.30
:Indy: indy , 150MHz R4400SC, 256MB RAM, XL24, 6.5.10
:Indigo2IMP: purplehaze , 175MHz R10000, Solid IMPACT
probably posted from Image bruce , Quad 2.5GHz PowerPC 970MP, 16GB RAM, Mac OS X 10.4.11
plus IBM POWER6 p520 * Apple Network Server 500 * HP C8000 * BeBox * Solbourne S3000 * Commodore 128 * many more...
could be worth the $200 in gas:
viewtopic.php?f=4&t=16728769
:Octane2: 400Mhz V8
surrealdeal wrote: could be worth the $200 in gas:
viewtopic.php?f=4&t=16728769


Mmm, no. That's a POWER3-II. Interesting as a collector's item, can do some basic tasks, might make a fun server to mess around with, but not a daily driver by any means.
smit happens.

:Fuel: bigred , 900MHz R16K, 4GB RAM, V12 DCD, 6.5.30
:Indy: indy , 150MHz R4400SC, 256MB RAM, XL24, 6.5.10
:Indigo2IMP: purplehaze , 175MHz R10000, Solid IMPACT
probably posted from Image bruce , Quad 2.5GHz PowerPC 970MP, 16GB RAM, Mac OS X 10.4.11
plus IBM POWER6 p520 * Apple Network Server 500 * HP C8000 * BeBox * Solbourne S3000 * Commodore 128 * many more...
I appreciate your post Cameron. Since I just became jobless a matter of a few hours ago, im calling off my search for now. One thing after a fucking nother eh?

I'm also sticking with my octane and ditching my Fuel because I just don't like it as much as my Octane. Plus as Hamei said, the Octanes are better when if comes to reliability. And I'm inclined to agree. The Fuel feels like an old x86 PC in terms of quality.

On installing NetBSD on an SGI, I'm not downgrading to an unaccelerated X server. I'm also not interested in any more apple products. So my options once I find new employment currently amount to late model Suns, late model HP PA RISCs and possibly an overpriced AEON or ACube system board. Seems like a paltry mix, but HP's PA RISC don't seem to be too much of a slouch even after they jumped to the Itanic, and I seem to be able to get a top of the line C8k for about $300.
:fuel: 900MHz 4GB
Wow, that sucks. I hope you can find new employment. I'm sorry to hear it. :(

Fuel vs Octane comes down to a matter of choice. I don't disagree the Fuel is "less special" with respect to design than the Octane, but I do love Big Red.

PA-RISC does have the advantage of cheap, but mostly because no one loves them anymore, including HP. They're pretty zippy systems, but my top-spec PA-8900 C8000 does use a lot of juice, and Linux is best described as a "work in progress" on it. At one stage I was thinking of porting Firefox 3.6 to it since it already does run Firefox 3.5, but I never got around to it and the Fuel replaced it in the KVM slot. It runs 11i v2 TCOE.

Like POWER, I have a soft spot for PA-RISC because my first job was on a K250 and later I did contract work with a C3750. It's a very clean RISC and HP at least crammed incredible amounts of L2 in it, something Apple could have learned from with their criminally undercached designs. I came to hate HP-sUX and yet become fluent in it. Now I have two HP-sUX machines (a 9000/350 in the huge tank-like minirack and the C8000), plus a "425t" that I need to figure out if it even still works and what the hell is in it.

If you can live with the C8000's limitations, and they're significant, it's certainly not a bad system.
smit happens.

:Fuel: bigred , 900MHz R16K, 4GB RAM, V12 DCD, 6.5.30
:Indy: indy , 150MHz R4400SC, 256MB RAM, XL24, 6.5.10
:Indigo2IMP: purplehaze , 175MHz R10000, Solid IMPACT
probably posted from Image bruce , Quad 2.5GHz PowerPC 970MP, 16GB RAM, Mac OS X 10.4.11
plus IBM POWER6 p520 * Apple Network Server 500 * HP C8000 * BeBox * Solbourne S3000 * Commodore 128 * many more...
ClassicHasClass wrote:
hamei wrote: Can you run Rhapsody on a G5 ?

No. I did get it running on this Wallstreet G3 and, well, it's NeXTStep with Platinum but no app compatibility.

Oiks. Thank you, just saved me some time. Mark that one off the "interesting" list :D
The time has come for someone to put his foot down ...
No, no, don't get me wrong, it's "interesting." The fact I have it on a Wallstreet laptop means I don't have to devote a KVM seat to it, I can just toss it on the project table when I want to play. One of these days I might make it do something practical, but the problem is that there weren't many NeXTStep applications then or now, and it doesn't have Carbon, so most Mac source code is out. OmniWeb was cool in the day but I'm better off running Classilla in Blue Box!
smit happens.

:Fuel: bigred , 900MHz R16K, 4GB RAM, V12 DCD, 6.5.30
:Indy: indy , 150MHz R4400SC, 256MB RAM, XL24, 6.5.10
:Indigo2IMP: purplehaze , 175MHz R10000, Solid IMPACT
probably posted from Image bruce , Quad 2.5GHz PowerPC 970MP, 16GB RAM, Mac OS X 10.4.11
plus IBM POWER6 p520 * Apple Network Server 500 * HP C8000 * BeBox * Solbourne S3000 * Commodore 128 * many more...
TeamBlackFox wrote: I appreciate your post Cameron. Since I just became jobless a matter of a few hours ago, im calling off my search for now. One thing after a fucking nother eh?

Aw, that bites. Here's hoping you get something better lined up before too long.

ClassicHasClass wrote: plus a "425t" that I need to figure out if it even still works and what the hell is in it.

425t is one of the 040-based Domain/HP-UX workstations. I just picked one up at the recycle center not too long ago, and it worked fine. Still need to get a keyboard/mouse or PS/2 adapter, though.
Computers: Amiga 1200, DEC VAXStation 4000/60, DEC MicroPDP-11/73
Synthesizers: Roland JX-10/D-50/MT-32/SC-55k, Ensoniq SQ-80/Mirage, Yamaha DX7/V-50/FB-01/SY22, Korg DW-8000/MS-20 Mini/ARP Odyssey/M1/03-RW, E-mu Emax HD/Proteus/2, Rhodes Chroma Polaris
No, I know what a 425t is -- I put it in quotes for a reason. Someone slapped a 715t/33 sticker on the front, but the backplate says 425t, and it does have the Domain/OS keyboard port suggesting the original motherboard. I just never get around to dragging it out of storage since my big 9000/350 satisfies my HP jones.
smit happens.

:Fuel: bigred , 900MHz R16K, 4GB RAM, V12 DCD, 6.5.30
:Indy: indy , 150MHz R4400SC, 256MB RAM, XL24, 6.5.10
:Indigo2IMP: purplehaze , 175MHz R10000, Solid IMPACT
probably posted from Image bruce , Quad 2.5GHz PowerPC 970MP, 16GB RAM, Mac OS X 10.4.11
plus IBM POWER6 p520 * Apple Network Server 500 * HP C8000 * BeBox * Solbourne S3000 * Commodore 128 * many more...
ClassicHasClass wrote: Like POWER, I have a soft spot for PA-RISC because my first job was on a K250 and later I did contract work with a C3750. It's a very clean RISC and HP at least crammed incredible amounts of L2 in it, something Apple could have learned from with their criminally undercached designs. I came to hate HP-sUX and yet become fluent in it. Now I have two HP-sUX machines (a 9000/350 in the huge tank-like minirack and the C8000), plus a "425t" that I need to figure out if it even still works and what the hell is in it.


PA's cache hierarchy is awesome - however, it's because of L1, not L2. As far as i know, no PA-8000-series chip has a true L2 cache. The 8800/8900 have external (off-chip) DDR DRAM L2 caches with on-chip tags, but performance is dismal; they're neither immensely high-bandwidth nor low-latency (over 40 cycles LTU - and higher for back-to-back access!).

That being said: PA-8500 and up have a superb large on-chip L1, with fairly low load-to-use latency. CPU people used to call PA-8xxx series chips "SRAM blocks that happen to have cores attached", which is basically true, and it's continued in the PA-derived Itanium family, which all have huge gobs of SRAM on-die (something like 75% of the Itanium 9300's transistor count is L3 cache!)

Nothing like cheap, low-latency loads to make a workload fly. :D
I guess it boils down to how you view it, but I'd still call that an L2 cache even if it's not as good as it could be. HP certainly did in all the spec sheets and it serves the same role.

EPIC/VLIW certainly needs huge cache for those instruction word sizes. :D
smit happens.

:Fuel: bigred , 900MHz R16K, 4GB RAM, V12 DCD, 6.5.30
:Indy: indy , 150MHz R4400SC, 256MB RAM, XL24, 6.5.10
:Indigo2IMP: purplehaze , 175MHz R10000, Solid IMPACT
probably posted from Image bruce , Quad 2.5GHz PowerPC 970MP, 16GB RAM, Mac OS X 10.4.11
plus IBM POWER6 p520 * Apple Network Server 500 * HP C8000 * BeBox * Solbourne S3000 * Commodore 128 * many more...
ClassicHasClass wrote: I guess it boils down to how you view it, but I'd still call that an L2 cache even if it's not as good as it could be. HP certainly did in all the spec sheets and it serves the same role.

EPIC/VLIW certainly needs huge cache for those instruction word sizes. :D


Yes, IPF is sensitive to cache - although not inherently because it's EP, but because it has an insane 41-bit op size plus a 5-bit template field for every three ops. This is the downside of having a massive number of GPR's, plus predication. :( For what it's worth, though, Itanium also has some density-friendly things too - for instance, not all instructions in a bundle have to issue concurrently, which reduces the need for NOP padding. In general, Itanium's cache is still a work of art (single cycle load-to-use from L1!) and performs magnificently.

Tilera's ISA is VLIW and puts either 2 or 3 ops (depending on the exact op type - some can only be issued as part of a 2-op bundle) into a 64-bit instruction word. As a result, its density is actually comparable to or superior to commercial RISCs.

Going to stop derailing this hilarious thread rsn...
I think I may dig one of the X86 boxes out of storage, with a fire extinguisher nearby, and set it up with Free or NetBSD and use it as a remote X server to bolster the limits of Mima. I also may get an Octane 2 from a friend who found one for free so that will help too.
:fuel: 900MHz 4GB
You could always get a zx2000 or zx6000 - they're pretty fast (by Old RISC Crap standards) and have good OS support.

Be advised, zx6000's are pretty loud.
Kira wrote: You could always get a zx2000 or zx6000 - they're pretty fast (by Old RISC Crap standards) and have good OS support.

Be advised, zx6000's are pretty loud.

Computing as a hobby is mort. You can play with old junk which is interesting but a dead end, or you can diddle yourself with whatever worthless crap the gnu fanboyz are in love with this week, or you can give up.

Truth is, home computing is now like Oakland : there's no there, there.
"move over theah, good buddy, cuz the Snowman is comin' through ..."
Hey, man, computers are much more fun as a hobby -- well, the old ones -- than they ever were as a career.

But then I'm collecting as a hobby the computers I used to work on when it was a career.

Now that I think about it, maybe you're right.
smit happens.

:Fuel: bigred , 900MHz R16K, 4GB RAM, V12 DCD, 6.5.30
:Indy: indy , 150MHz R4400SC, 256MB RAM, XL24, 6.5.10
:Indigo2IMP: purplehaze , 175MHz R10000, Solid IMPACT
probably posted from Image bruce , Quad 2.5GHz PowerPC 970MP, 16GB RAM, Mac OS X 10.4.11
plus IBM POWER6 p520 * Apple Network Server 500 * HP C8000 * BeBox * Solbourne S3000 * Commodore 128 * many more...
ClassicHasClass wrote: Hey, man, computers are much more fun as a hobby -- well, the old ones -- than they ever were as a career.

I can't talk about computing for computing's sake but as a means to make things, computing in the seventies wasn't exactly fun so much as, "Wa ! Can you believe I just entered that entire program in thirty seconds ! If I had to type it in it would have taken three hours and had fifty mistakes !" A serial connection to the teletype was waay cool !

The PDP-8's that ran K&T mills still had to be toggled through the bootstrap procedure to enter their executive program. Good thing they had core :)

And when they screwed up, there goes $3,000 to repair that 32k core board. That's in 1975 dollars ...

But an NC machine was ten times faster than a human with better quality. Even if the steenkin computers drove you crazy, at least there was some gain. So when you spent your time and money on some "new and improved !" doo-wacka-doo, at least you felt like you were getting something.

(If I had it to do over again I would not spend ten cents on any of that crap. Today you could make a better living doing onesy-twosies by hand, since there are so few places who can do that now.)

As a hobby or an interest, computers were nifty because you could see almost endless potential. OpenDoc. Display Ghostscript. The Workplace Shell. SMP. Threading. Out of twenty different choices, there was going to be something you'd find suitable.

Now what do we have ? javascript ? dynamic html to push advertising swill down people's throats ?

Shit, that's what we have. GNU has become the monster it hated while the commercial enterprises have become what monopolies always do : "You will buy what we want you to buy at the company store, or die. Good luck, sucker. Drop off your cash on the way out, would you ?"

If this was the cost of progress, well, maybe. But what progress ? There hasn't been anything of value happening in personal computing for a good ten years.

Discouraging :(
"move over theah, good buddy, cuz the Snowman is comin' through ..."
@ CommodoreJohn:
If you find someone doing a prod run of the HIL<->PS/2 please let me know. I've got a 425T I'd like to get doing some useful things. Not sure if it's stuck in domain mode. :)
Al Boyanich
adb -w -P "world> " -k /dev/meta/galaxy/ksyms /dev/god/brain