The collected works of jan-jaap - Page 14

On e.g. IRIX there's a speed penalty to 64bit: pointers are twice as long so you'll use more memory, run out of CPU cache faster etc. On these systems you go 64bit only when you have to because your application requires > 2GB RAM.

With x86-64, something different is going on: with 64bit support came various ISA extension which make that 64bit code is usually faster then 32bit equivalent.

_________________
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
ClassicHasClass wrote:
With the KVM, the machine starts fine, but if I switch away the PROM monitor freaks and reports a message "usb_hid_reattach" or some such.

FYI, I've got a Fuel attached to a Belkin SOHO USB/DVI KVM and that works just fine. Or should I say, worked just fine, the Fuel's DALLAS ran out and with a garbled NVRAM it can be quite cranky when I power it up.

_________________
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
ClassicHasClass wrote:
Memory is hard to find for those units. I've learned to be satisfied with 16MB (which is still a lot for HP-UX of that generation).

Not my auction. Wrong continent. IMHO quite pricy (price is per stick). Still: http://www.marktplaats.nl/a/computers-e ... 04-56.html

_________________
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
robespierre wrote:
The 3600 is a real beast, it makes a predator rack look cute.

I object to that. The Predator rack is cute, period. :mrgreen:

Can't wait till the wiring of my new room is done, then I'll fire it up again 8-)

_________________
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
Late rev. V10 cards shipped with 128MB RAM. The possibility to unlock this and turn it into a V12 was discussed here . AFAIK, it is impossible to rewrite the chips that hold the part# and XIO ID string, (e.g., in your case ASTODYV10 and 030-1826-00x).

In short, it looks to me that someone went ahead with the info in the thread mentioned above and figured out how to turn rev B. V10's into 'fake' V12s and didn't feel the need to share this info with the board. :x If I were you I'd feel scammed, because this is NOT a genuine V12. The part# clearly makes this a V10.

_________________
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
Took me a moment to figure that one out :)

Have you seen this guy's work? http://www.dailymail.co.uk/news/article ... hotos.html

_________________
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
I ran an AXPpci33 as a firewall for many years. Those systems used x86 emulation to initialize the ROM of the video card, and IIRC, that didn't play nice with many video cards. If it hangs just after the 'BIOS emulation v2.03' message, this may be the problem. Tried a different video card yet? I think I used a Matrox Millenium (with VRAM option card 8-) ) in mine.

_________________
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
Same problem here -- I've got several SDI/HD-SDI equipped SGI systems but no decks, monitors etc. I've got a VideoLab in my 4D/380VGX with parallel digital (601) I/O and a pair of AJA SDI converters. In theory at least, I think that PowerSeries can do some of the tricks that the O2 can do, like mapping a live video stream as a texture on (moving) geometry. I could feed it with SDI video from my Onyx2 or Tezro. Now that would be some excellent retro-computing fun :-)

On the plus side: the 'traditional' mini converters from AJA and the like still cost a fair bit of money, but it looks like the Chinese have jumped in, e.g. http://www.ebay.com/itm/350592200330
They may either be good (or at least reasonable, I'm not doing broadcast stuff anyway), or else drive the price down.
To accentuate the special identity of the IRIS 4D/70, Silicon Graphics' designers selected a new color palette. The machine's coating blends dark grey, raspberry and beige colors into a pleasing harmony. ( IRIS 4D/70 Superworkstation Technical Report )
The DM10 is a PCI card. Just a generic FireWire card with a DM10 sticker, to be precise :)

The device driver doesn't care about the manufacturer or the color of the PCB. It cares about the PCI vendorID:deviceID of the chip(s) on the board. In IRIX, a device driver can register itself as a handler for one or more PCI IDs using the function pciio_driver_register() . It usually does this in the entry point ${DRIVER_NAME}_init().

It wouldn't be kind to Nekonoko to post disassembled IRIX code here, but if you run fire.o through dis and have a look at fire_init(), you'll see it calls pciio_driver_register() three times, for these PCI IDs:

The Adaptec board recognized as a DM10 has a PCI ID of 104c:8024.

So, it is to be expected that any card with a PCI ID of 104c:8024 is recognized as a DM10. The other two IDs are probably a courtesy of the driver writer because SGI didn't ship products with those IDs. These appear to be recognized as '1394 high speed OHCI card'.

The IRIX kernel will not call the 1394 driver for any devices not registered by it. This includes your FW800 Adaptec, regardless of it's shiny red appearance ;)

TI made quite a few different 1394 link layer chips (and corresponding PHYs), but the three mentioned here are the most common in consumer equipment. Hence Hamei's sentiment that 'anything with a TI chip' works (even if the PCB is green :mrgreen: )

Those handy with hex editors might get creative and patch the PCI IDs supported by the driver. I'd be careful with that, though: several 1394 chips, especially VIA chips, have really weird quirks. An 800Mb TI chip might work, though. Maybe I'll give it a shot.

And that, boys and girls, concludes today's introduction to reverse engineering :lol:

_________________
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
OK, I couldn't resist. Put a MacPower FireWire 800 card in the Fuel , and patched the link layer PCI ID in (104c:8025).

It is picked up, and reported in hinv:
Code:
1 500 MHZ IP35 Processor
CPU: MIPS R14000 Processor Chip Revision: 2.3
FPU: MIPS R14010 Floating Point Chip Revision: 2.3
Main memory size: 1024 Mbytes
Instruction cache size: 32 Kbytes
Data cache size: 32 Kbytes
Secondary unified instruction/data cache size: 2 Mbytes
Integral SCSI controller 2: Version IEEE1394 SBP2
Integral SCSI controller 0: Version QL12160, low voltage differential
Disk drive: unit 1 on SCSI controller 0
Disk drive: unit 2 on SCSI controller 0
Integral SCSI controller 1: Version QL12160, single ended
CDROM: unit 1 on SCSI controller 1
IOC3/IOC4 serial port: tty3
IOC3/IOC4 serial port: tty4
IOC3 parallel port: plp1
Graphics board: V10
Integral Fast Ethernet: ef0, version 1, module 001c01, pci 4
IEEE 1394 High performance serial bus controller 0: Type: OHCI, Version 0 0
USB controller: type OHCI
USB Human Interface Device: device id 0 type keyboard
USB Human Interface Device: device id 0 type mouse

(notice the IEEE 1394 High performance serial bus controller and Integral SCSI controller 2: Version IEEE1394 SBP2 )

*but* at boot it mutters something about a loss of SCLK and unable to reset the bus.

I happen to know a thing or two about FireWire (cough) and this isn't good. Basically the interface between the discrete link layer (TSB82AA2) and PHY (TSB81BA3) chips isn't properly initialized. The TSB82AA2 is a 1394B OHCI 1.1 link layer, and that's too different from 1394A OHCI 1.0 chips to 'just' work without changes to the code. Additionally, the TI TSB81BA3 PHY has a bug described here: http://www.ti.com/lit/er/sllz059/sllz059.pdf which needs to be worked around or it simply won't work.

And indeed, an attached Lacie FireWire 800 hard disk doesn't show.

OK, now back to work ;)

_________________
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
pentium wrote:
There was chatter in the IRC the other day that when seen as a DM10 you can attach DV devices like MiniDV cameras and decks and both stream and control to whatever software you have installed. [...] That's a hell of a lot more than just hanging disks off the system.

All true.
pentium wrote:
It IS worth nothing that my Macally card (and possibly the Adaptec card too) has an "EEPROM WRITE PROTECT" jumper. I'm curious if you can change the Device ID to something Irix would be more expecting.

Whoa there. There's a *big* difference between changing the sub VendorID: sub DeviceID (the essence of 'fixing' a 3c996bT or other OEMed TG3 gigabit card) and changing the vendorID:deviceID of a card.

The subVendorID:subDeviceID exists to allow the likes of SGI to sell an otherwise identical $10 IEEE1394 card as a $1000 video option.

The vendorID:deviceID exists to uniquely identify different, potentially incompatible, pieces of hardware. *How* incompatible is anybody's guess.

If *I* wanted to do this, I'd try a different approach. This is all documented, you know . The mechanism of a PCI driver roughly works like this:

1) The fire_init() function register the driver as the handler for the three device IDs mentioned yesterday.
2) The IRIX kernel scans the PCI bus(es) when booting, and if it encounters PCI devices, it attaches the registered driver for every instance it finds. If nothing is registered for a PCI ID, noting is attached.
3) So fire_attach() is called if any of the three device IDs is probed. But we've seen it doesn't *act* the same for each of these PCI IDs. In other words: these likely a secondary check in fire_attach() which checks the PCI ID for which it's being called, and depending on the results, registers either a "high performance IEEE1394 bus" or a "DMediaPro DM10 FW option" vertex in the hardware graph (the pseudo-fs under /hw).

So, to turn one of the two devices (device IDs 0x8009, 0x8019) into pseudo-DM10's, you'd have to locate the secondary check for 0x8024 and patch it out, or change the check to match your device.

And if your time is worth anything at all, you spend $10 on a card with the right PCI ID and it just 'works' (with all the limitations and quirks of the original DM10)

_________________
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
robespierre wrote:
I used to use Altsys's Metamorphosis Pro on my classic Mac. It works very well.
http://vimeo.com/14758980

I used it's predecessor (Fontographer) years and years ago.

I liked it a lot.

_________________
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
hamei wrote:
jan-jaap wrote:
pentium wrote:
There was chatter in the IRC the other day that when seen as a DM10 you can attach DV devices like MiniDV cameras and decks and both stream and control to whatever software you have installed.

All true.

J-J : are you saying someone somewhere has got this to work in Irix or that it is all supposed to work in theory ?

I have personally used my Panasonic mini-DV camera (NV-GS220), a FireWire webcam and a Lacie FireWire disk with a 'fake' DM10 (random 1394 card with the right [104c:8024] TI chip). I could crash the O2 running 'fsr' with a 1394 disk attached initially, but at some point that was fixed. The lack of hot plug makes it rather unusable though.

I am aware that people have had troubles with various FireWire devices over time, but how much of that was due to non-genuine DM10's, problems with early releases of the DM10 software? I don't know...

But my last message was more about the difference between cards with 104c:8009, 104c:8019 IDs and the 'DM10' (or compatible), the 104c:8024. The latter has the DigitalMedia functionality, and my post was about ways to (in theory) unlock this functionality on the 104c:8009 and 104c:8019 devices.

That was all speculation, but that's how I would do it. I had a quick look at fire_attach() and indeed it does the equivalent of

Code:
pciio_device_id_t device_id = pciio_config_get(vhdl, PCI_CFG_DEVICE_ID, 2);

if (device_id == 0x8020 || device_id == 0x8024) {
/* do DM10 things? */
}


0x8020 is a TSB12LV26 IEEE-1394 Controller (Link). Sloppy, the device driver has not even been registered to handle this device ... but I'm pretty sure if you patch 0x8009 or 0x8019 in there it will unlock DMedia functionality on those boards :mrgreen: I don't think I have either of these two boards, so I'll leave this for others to figure out.

And that, boys and girls, concludes today's reverse engineering tutorial.

_________________
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
canavan wrote:
Apparently, there's no magic going on in there, otherwise Discreet would not have been able to offer an external DVI splitter (a not SGI specific product made by Gefen) as an alternative to the DCDs, nad apparently, there are even cheaper alternatives, like the one offered here viewtopic.php?f=4&t=16727045

Well, if you configure the DCD to duplicate a single screen across both DVI connectors, then you can functionally replace it with a DVI splitter (such as the one sold by Gefen)

But if you configure the DCD to two independent screens (the 2@ modes), then I don't see how you can replace it with a splitter.

I think you've just proven that Discreet setups don't use 2@ modes :? Somebody with a working setup should be able to figure this out ...

_________________
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
hamei wrote:
zeppelin crashes

Good suggestion. Now playing :)

_________________
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
mila wrote:
How tall are these cabinets?

185cm

_________________
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
I've got a QLA2342 in my O2 . That's a dual 2Gb/s FC card capable of close to 400MB/s throughput (with a suitable array attached). First of all, the 64bit, 33MHz PCI bus of the O2 limits this to ~ 200MB/s. But real (diskperf) numbers are well below that, IIRC 60-70MB/s.

There's also an LVD version of the Adaptec 2940UW, the AHA-2940U2/U2W. I've had one of those in another (195Mhz R10K) O2 at some point. Requires a bit of PCI id hacking in the driver (similar to using an Adaptec 3940 ), but will provide similar performance.

I think the CPU speed doesn't even matter in the end, ultimately the O2's unified memory architecture 'bus' is where the limitation is.

Winnili wrote:
How easy and practical, or even possible, would it be to add an Ultra-3/Ultra160 HBA inside the O2 and to which degree would the O2's overall bandwidth deal with it acceptably? And which of which, if any, work inside the O2?

Ultra-3 is LVD. Roughly speaking, SE SCSI (what the O2 has) requires 16 data lines plus ground, and LVD requires 16 differential pairs. So, this would require a redesign of the O2 mainboard and backplane. If you're going to do this, would you please raise the performance limitations of the chipset and allow for faster CPUs' and more RAM while you're at it? :mrgreen:

_________________
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
Unless if you have a fetish for SX (optical) networking, I'd go with a Broadcom 5701 based card such as the HP NC7770 for any system with a PCI bus (except the O2).

Requires a bit of hacking of driver or card (all documented here), but they run well, they run cool, and they're on eBay for as low as $2.45

_________________
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
Winnili wrote:
I meant more by adding some type of PCI bus expansion card.

A suitable 2Gbit FC card like the QL2342 with a small enclosure for 1 or 2 disks is probably the easiest solution because it requires no hacking. It is a fully supported way to attach a disk system capable of transfer speeds the O2 will never be able to utilize 8-) . But you won't be able to boot from it because the O2's PROM can only boot from an Adaptec 2940.

If you want something bootable, and you were able to hack the PCI ID of an AHA-2940 into a 2940U2/U2W you might be able to boot from an (external) SSD-with-ARS-2160H and get the combination of the lower access times of the SSD, and the higher transfer speeds that the 2940U2/U2W is capable of.

Anything beyond that mostly depends on your ability to write drivers for it. That, and it has to support 5V PCI I/O of course.

Personally, I think it's all a bit of a waste of energy. Even an entry level Octane (or even an R10K Indigo2 IMHO) is a snappier, more responsive system than the O2. I'd like to get my hands on one of those GIO64 FC cards, see how far you can push an Indigo2 :)

_________________
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
Nice :)

recondas wrote:
Currently running a virtual 3200x2400 desktop.

Let me get this straight, you're not using the Compositor in any way in this setup, right?

_________________
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
Did you read this thread: viewtopic.php?f=3&t=7894 ?

Do you get any kind of prompt when you power up the system ('pwr u') after it print the message about disabled CPUs?

You may have to set the virtual dip switches to boot the system into POD or CAC mode instead of booting the regular PROM. Then, from there, the 'clearallogs / initalllogs / flush / reset' sequence mentioned in the link above (and other places on this board).

Sorry, I don't have a link handy to the virtual debug switch documentation. If nobody can help you with those I can dig a little deeper.

_________________
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
Excellent, have fun!

_________________
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
Hi & welcome :)

computron wrote:
i have bought a crimson last week end

Red fever made another victim :mrgreen:

_________________
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
I've never seen a bad contact between a CPU and mainboard.

If the CPU module goes bad, it's almost always the cache memory, and IRIX should complain about uncorrectable cache errors.

I have to clean the memory of my Indigo2's every once in a couple of years. Especially the R10K Indigo2 with 1024MB RAM because the 'gigaram' chips don't have gold contact fingers.

Validate your cleaning with the diagnostics tests (press ESC to interrupt the boot process when prompted and select 'run diagnostics).

_________________
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
Feelies wrote:
Also, I have seen some very high pricings on the Sun models. I understand that they (and the rest of the computers on my list) are relatively recent machines, but all of the Sun Ultra 45's on ebay are listed at several grand or more, while I have seen listings for most of the others for under 1000. Is this a sane price for these?

Isn't the Ultra 45 the last of the Ultra workstations? In that case it's like the Tezro which are also very expensive.

Here's one (not my sale): http://www.marktplaats.nl/a/computers-e ... ra-45.html

Asking price is 650EUR. No idea if that's a sane price :)

_________________
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
hamei wrote:
The Composer goes in the other direction. It accepts several inputs to output an equal-or-lower resolution, higher-memory display. It won't go to 1920 x 1200, either.

Why not?
compositor manual wrote:
The compositor features include the following:
• Four DVI single-link digital video inputs with up to 165-MHz pixel clock per input can combine into one digital video output that allows up to 165 MHz of output.
– A selectable pixel averaging feature (antialiasing) combines the inputs from up to four graphics pipes into the compositor and averages these inputs to smooth jagged edges on displayed images.
– A spatial composition/tiling feature enables each pipe connected to the compositor to render a pre-defined portion of the display

165MHz output ( single link DVI ) should be capable of 1920x1200 ?

_________________
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
I've ported GCC3 to IRIX 3.3: ftp://ftp.jurassic.nl/pub/irix/gnu/irix-3.3/

The problem with IRIX 3.x is that the C library is pre-ANSI so it lacks a lot of functionality. This also makes porting existing code to it a lot of work. Not even mentioning the slow CPUs of these systems and that e.g. GNU 'make' isn't exactly something that makes good use of the 3D hardware or that you haven't seen on the nearest Linux system.

_________________
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
Thinking about starting a collection

LOL, a collection is not something you plan, it's more something that grows organically. Like a virus :lol:

I think it starts for most people because they like to own something they used to work with when it was new, amazing and completely out of their (financial) reach. That's how I got my first Indy back in the nineties. You mess with it, if you like it you buy another, at some point you run out of space and sell the things that don't appeal to you anymore. Or you have to narrow your scope -- at some point I owned PA-RISC, SPARC and AXP hardware and a lot of SUN storage -- now it's only SGI (OK, I still have an IBM DS4000 SAN and a Dell PowerVault).

At least that's what happened to me -- and I own a fairly substantial collection of SGI's :)

So, go ahead and buy that one thing you always dreamed of owning. Then see where it goes from there. I hope you have some friends or better half to tell you when enough is enough :lol:

_________________
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
hamei wrote:
Did 3.3 have Motif ?

I don't think so. It has some ancient X11 runtime (X11R4?). I've got the old v1.2 Motif sources, but the X11 development system was an option for IRIX3 and I don't have it :(

A sufficiently old X11 should build from source (the libs, not the X server). I tried that once, it had IRIX (or was it IRIS?) specific code, but it doesn't build out of the box.

Remember that shared libraries are a royal pain in the behind on pre-ELF IRIX. The process is rather involved, will almost certainly have to be done manually. You have to hard-code the load address which may not overlap with another shared library. The same mess as pre-ELF Linux, basically.

_________________
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
hamei wrote:
I have seen hints here and there that with the latest, largest memory cards it might be theoretically possible.

I doubt that. The problem is not the amount of memory on the raster manager but the bandwidth of the pixel bus to the DG5.

OTOH, if you lower the refresh rate you bring down the bandwidth consumption and it might work.

_________________
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
canavan wrote:

Code: Select all

NODE        IP59_4CPU       NTR758         030_1989_003          C    00
If that one is yours: congrats. The 4x1GHz Tezro is about as good as it gets :)

Looks like this one?
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
canavan wrote: No, it doesn't look like a deskside Onyx2. I thought you'd know that...

copy-paste fail :oops: Let me try again: viewtopic.php?f=14&t=12569 ?

Both are purple and highly desirable, though :mrgreen:
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
jpstewart wrote:
I'd love to know how the Chinese price list compared to one of a similar vintage from North America. I wouldn't be surprised if those prices were 5 to 10 times higher than the rest of the world.

Spot on! One year later (March 2005), US pricing before discounts:
Code:
WF-700V12-1036 Silicon Graphics Fuel V12 Graphics 700MHz
R16000/4MB Cache, 1GB Memory,36GB System
Disk. Order Monitor Separately
$14,595

It's $71,762 in hamei's list ...

But it's the big iron where things really hurt, e.g. in 2006 you paid $25K for a GE16-4, $45k - $55K for an RM etc etc.

_________________
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
If you've got a SCSI backplane you've got an IO9: http://techpubs.sgi.com/library/tpl/cgi ... 631-PARENT

Yes, it has SCSI and IDE, but they are on different connectors.

pic2.jpg
pic2.jpg (65.21 KiB) Viewed 630 times


(ok, these are Origin 350, but the Altix is similar)

You need power to the backplane too.
To accentuate the special identity of the IRIS 4D/70, Silicon Graphics' designers selected a new color palette. The machine's coating blends dark grey, raspberry and beige colors into a pleasing harmony. ( IRIS 4D/70 Superworkstation Technical Report )
What I find most offensive is you have pay import tax on the value of the item plus shipping . If you import something from the other side of the planet, shipping cost (and therefore import tax) can be higher than the cost of the item.

IMHO it is just not fair that my (Dutch) government is taxing the work of foreign people processing my package.
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
dbx is on the ProDev WorksShop disc. The last ProDev WorksShop disc (2.9.5) has dbx 7.3.7. Patch SG0005926: DBX 7.3.7 Debugger fixes rollup #1 applies.
Code:
janjaap@speedo:~$ dbx
dbx version 7.3.7 (96413_Dec19 patchSG0005926) Dec 19 2005 14:42:01
Debugger Server version Dec 19 2005 14:44:41
(dbx) quit

It's not nodelocked. Install & enjoy :)

_________________
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
I had that with an Octane MXE board. Reseated the TRAMS and the problem went away.
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
Small update. I spent a good amount of time arranging infrastructure stuff (19" rack, switches, SAN stuff etc etc.) Now I can start putting in the wiring to the systems. Networking, fibre channel and the odd serial cable. So I've started to sort of move them towards their final place and build up the Ikea shelving system that will hold most of the workstations. The Ikea Broder series has been discontinued but I managed to get one second hand for a decent price.

I needed a PDU for my 19" rack and got a pair of APC AP7953's . One is in the rack, the other is under the floor behind the SGI desksides. This way I can (remotely) supply power to the desksides without using the breaker switches on the back of them. First of all, that's not what a breaker is for, second they are kind of hard to reach without moving the systems around. I will wire the workstations to the APC PDU as well. This way I can be sure no power is supplied to any system I'm not using. Most SGI's use anywhere between 5 and 20W of electricity when plugged in (even when powered down). If you've got 15 of them that starts to add up. This is better in terms of fire safety as well, as many of these systems are now 20+ years old.

I'll probably end up using 3 or maybe 4 shelves for the Ikea Broder rack. Put the Fuel and the Indigo2's underneath, Indigo's, Indys, Octanes, O2 etc. and some misc small SCSI gear on the shelves.

The panels of the 19" rack are against the back wall for now. That's roughly where the book cupboard will be for all the manuals. Above it is the drain for the airco.
room_720_1.jpg
Deskside parade

room_720_2.jpg
Test fitting
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
Thanks for all the kind words guys

mia wrote: I have more like a vertical setup.

I like it. What's that storage you've got there?

I can do vertical too, it's on the other side of the room :) I still have to wire (almost) everything and move the server. I reserved space for my Origin 350 (currently hosted at $WORK) too. The GS724T switches are the biggest you can buy without active cooling. Now if only I could shut up that bloody Silkworm, it is absolutely intolerable....
DSC_2077.JPG
4D/380 VGX and infrastructure rack

storage.jpg
TP9100, IBM DS4300+EXP710, Dell PV220S

switches.jpg
Cisco WS-C1400, sanswitch, twice GS724T and a 48port serial concentrator


Pontus wrote: I've been eyeing the "broder" shelf system but I'm not sure about the grating. Will you put something smoother on top of that, to save on the computers rubber feet?

I didn't think of that beforehand. I've got one or two normal shelves (without the grating) but somehow they aren't very nice. I think I'll either put pads under the systems, or some sort of placemat on the self. We'll see.

Oh, and I finished the upgrades to the alarm system. Fire *and* burglary. I had the wires put in during construction, I just didn't have the time before to program the new zones into the system. No pets allowed in this part of the house, so the sensors are set to maximum sensitivity :twisted: I'll just have to make sure the thermal footprint of the larger systems doesn't trigger the alarm :lol:
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)
Pontus wrote: I've been eyeing the "broder" shelf system

Some things worth sharing about the Broder series. What I wanted from a shelving system was:

* Something that would last me a long time. So nothing with chip wood shelves for example.
* Not something with legs on all four sides because the back legs would rest on my cable gutter.
* Something 60cm deep. Most shelving systems are designed with archiving in mind and are 30 - 40cm deep.
* Something with a 'metallic', 'industrial' look to make the colors of the workstations stand out.

Enter the Broder series. That was ~ 1 year before I started construction of the annex and I didn't want to buy it back then because I didn't see the point of spending money on something, only to have to spend more to put it in storage for another year. Of course by the time I went to the Ikea to buy one they had stopped selling it. :( A quick search on the internet will show that many people are unhappy about this, and second hand prices are close to or sometimes even higher than the new price. It sells quickly second hand, and due to the wide range of configurations chances of finding exactly what you need are not good. I started looking for alternatives but there aren't many. You quickly end up in shops selling store inventory and price tags are *not* hobbyist friendly...

In the end I found someone who was selling a large lot of Broder parts, and had more or less mislabeled them so they wouldn't show up in the searches. I bought the lot. My car (VW Passat Station) was packed front to back, and from the floor to the roof 8-) I will eventually sell off the parts I don't need and hopefully reclaim some of the money I spent.

I'm using the 60x120cm grated shelves, and the 60cm 'T' legs. The legs (optionally!) came with adjustable, spring loaded extension to bolt them to the ceiling. I highly recommend these to keep the rack from tipping over.
DSC_2088.JPG
Broder leg extensions

One thing I had not anticipated was that the tolerances of the parts are rather 'loose'. The extensions I mentioned will rattle if you shake the legs for example. I was afraid that a running computer system would cause nasty vibrations so I applied duck tape in the right places.

Another place where the tolerances are annoying is the brackets that attach the shelves to the legs:
DSC_2087_01.JPG
Broder shelf attachment

DSC_2085.JPG
Broder brackets

You can imagine that if the hooks are even slightly too far apart it will allow the bracket to 'rotate' on the vertical stand. A rotation of even 1 or 2 degrees will result in the shelf moving 1 cm up or down at the edge -- unacceptable. Judging by the tape remnants everywhere, the previous owner had struggled with this as well. Now I'm happy I've got this ridiculous amount of parts because I can select the best, and then, using hammer and pliers make them just right. I hope it will stay solid for years to come, but only time will tell...

Oh, and yes, one of the three feet of each 'T' leg is still resting on my cable gutter. I think the bulk of the weight is on the center foot and the outer feet are mostly for balance. But I've added some reinforcements inside the cable gutter anyway. What's more annoying is I can't open the gutter without removing the rack first. I will probably cut the lid of the gutter in a couple of places to be able to access the gutter under / behind the rack. That's why this is still not fully assembled, or bolted to the ceiling. I have to be absolutely sure before I cut the floor panels on the cable gutter ...
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Onyx2: (2x) :O3x02L:
In the museum : almost every MIPS/IRIX system.
Wanted : GM1 board for Professional Series GT graphics (030-0076-003, 030-0076-004)