Sonnet 7200 PCI programming - Page 88

grelbfarlk · 14 February 2018, 00:14

Quote:

Originally Posted by Cego

btw, should i upgrade my PSU after installing the Sonnet G4? I am also planning on using an 060 turbo board with scsi and replace my harddisk with an SSD or CF card. There are no other Zorro cards, just the Voodoo and Realtek.

In a properly working* system the standard A4000 power supply should work.

*Few systems are properly working anymore

The other concern is whether the 3A VRM on the Black Mediator is enough for the G4, it might be.

Wrangler · 16 February 2018, 15:42

Update on the Frankenfest (A500+/ACA500+/Mediator): the mediator autoconfigs, PCIINFO recognises the Voodoo card but any attempt to open the Voodoo monitor driver causes a guru of various different types. It seems close to working but my suspicion is that there is a memory problem - perhaps lack of chip mem on A500 compared to A1200? Shame because it was looking both unbelievable and like it was going to work.

trixster · 16 February 2018, 17:09

I got gurus when opening the voodoo monitor driver when I didn't have layers.library v45 installed correctly. Are you sure p96 etc is setup correctly?

Wrangler · 16 February 2018, 18:33

I'm pretty sure the Picasso install is good and I reinstalled it for good measure with the same results. I'm running this with an '030 - that wouldn't make any difference?

grelbfarlk · 16 February 2018, 23:14

'030 should be fine but if you're using A1200 Mediator you might need to play with the MMU variable if it exists or not. P96 I think requires a minimum of 020. Also make sure you're not using all of the 8MB of Zorro-II space maybe?

Wrangler · 17 February 2018, 21:59

Sadly, no amount of fiddling around with 1200TX mem window jumpers, MMU envs (and all the others), ACA500+ parameters etc makes any difference. I think this is a simple case of a step too far with hardware that was never designed to work together.

nexus · 18 February 2018, 12:24

do you have latest firmware for the aca500+?

it uses a new MMU library

http://wiki.icomp.de/wiki/ACA500plus
might need to play with those settings also

http://wiki.icomp.de/wiki/ACA500plus_registers

Software-defined autoconfig

This feature helps getting drivers into the Amiga OS without the need for ugly hacks. With some A1200 accelerators "covering" the $f0.0000 area, the old method that was used on the ACA500 had to be improved. The solution is to sacrifice 64k of memory (of which "plenty" is available) and prepare it with autoconfig data. Although any vendor ID and product ID can be used, you should use the iComp vendor ID 4626 ($1212) and product ID 65 ($41) for the autoconfig nibbles. These have already been added to the latest boards.library.

Since there is no config_in and config_out lines on the A1200 accelerator port, the ACA500plus implements a "sniffing circuit" that will generate these signals internally. For this circuit to work properly and maintain the config chain, the system must know ahead of time which address the 64k-sized board will be mapped to, and how many Autoconfig boards are located within the A1200 accelerator. A few examples:

ACA12xx accelerators use no autoconfig at all. The ACA500plus should be programmed to be the first autoconfig board.

ACA1233n accelerators use Z3 autoconfig for it's memory, which will end up outside the Z2 area. The ACA500plus should be programmed to be the second autoconfig board and to final address $e9.0000.

ACA1221 accelerators use Z2 autoconfig for a 64k-sized IO space, which will be mapped to $e9.0000. The ACA500plus should be programmed to be the second autoconfig board and to final address $ea.0000.

Blizzard 1230-IV accelerators use Z2 autoconfig for a 128k-sized IO space, which will be mapped to $ea.0000. The ACA500plus should be programmed to be the second autoconfig board and to final address $e9.0000.

Joe · 22 February 2018, 01:13

Hello everyone,

I would like to inform about my sonnet issue:

- My sonnet was tested by Jack3D and working
- My RAM which I borrowed to Jack3D one piece suddenly began working on last time when he tested
- Jack3D sent back to me sonnet and I tested with new RAMs which I ordered and with the same result - Ramlib Error after inicalizing powerpc.library
- I bought Voodoo 3 3000 and I replaced the gfx card. I booted Jack3D system and after initializing powerpc.library the same error was occured - RamLib Error

So, already I see the RamLib Error if I use Radeon 9200 gfx or Voodoo 3 3000 gfx. So, I do not know where the problem might be. Jack3D told me that I have strange mediator. My mediator working fine without sonnet or with sonnet but without inicializating initppc command.

My mediator was produced in 2001 year. I see that working correctly.
What do you thing where can be a problem?

BTW what does it mean this error - RamLib Error 80000004 , I tried some patch tools as PatchRam, MCPRAMLibPatch or etc. but without successfull.

And one more thing. What error will be showed when you added bad RAM in your cause. What error
was showed like in my cause or other error?

Thank you very much for your tips.

grelbfarlk · 24 February 2018, 00:19

I'm not understanding a little bit.
You had a stick of RAM which didn't work for you but suddenly started working for Jack.
The same board with the same RAM didn't work when you got it back?
But with totally different RAM which should work, it doesn't.

Like I was saying in the private messages, usually RamLib 80000004 is a 68k crash. So what else is different between your system an Jack's? Is he using a custom 3.9 ROM and you are using original KS3.1?

Also as I said in the PMs here are some of the library error messages which have been associated with but not necessarily caused by memory issues:
"No memory detected on the Sonnet card"
"Memory corruption detected during setup"
"PowerPC CPU possibly crashed during setup"
"PowerPC CPU not responding"

Also do all memory slots work the same? Depending on revision on the CPU side of the board there should be small SMT resistors that were added later to the back of the memory slots, I've had two boards which one of the resistor legs were not soldered just right. Caused sometimes memory to show up or crash when Initing. If you are able, very gently try to check the resistance of them on the pins of the socket and maybe very very gently try to wiggle it slightly to see if the connection is broken.

Don't know if you did this but if you have the one old DIMM working try just that one first. Then replace with just one new DIMM.

jack-3d · 24 February 2018, 01:50

Simply he sent me his sonnet and ram. I checked and it was not working. I replaced ram by my module. It was working fine. Last day just before packing I tested his ram again, just to make screenshot of the error and suddenly his ram was working (even after few restarts). Miracle?

Anyway I send him back card, ram and also disk with my system fully tested. But he still has the ram lib error.

So...
Sonnet = tested, working
RAM = tested, one day not working, next day working
OS = tested, working
Voodoo = not tested
mediator = not tested
mainboard = not tested

I would bet on mediator.

nexus · 24 February 2018, 02:36

sounds more like a weak solder joint on the sonnet ram slots

not sure how you guys are setting them up but its easy to flex the sonnet while inserting and if ram is already installed then it could be rocked out of the slot so after installing make sure its re-seated and clean

fyi you shouldnt mix ram that use different plating on its contacts
eg
gold and tin
if the slots are tin use tin plated ram sticks vice versa for gold

Wrangler · 24 February 2018, 20:46

A fresh OS install cured this problem: now I can attach the Voodoo board in P96Mode. But trying to open a Voodoo screen hangs the system. ScreenMode puts the normal Amiga screen to white/blank (as you'd expect) but the Voodoo does not generate a signal, whereas P96Mode just hangs. Either way the Voodoo doesn't get hot. CheckBoards or whatever it's called in the debug folder seems to find the card with correct mem address etc. Before anyone suggests, I have reinstalled P96...

Quote:

Originally Posted by Wrangler

Update on the Frankenfest (A500+/ACA500+/Mediator): the mediator autoconfigs, PCIINFO recognises the Voodoo card but any attempt to open the Voodoo monitor driver causes a guru of various different types. It seems close to working but my suspicion is that there is a memory problem - perhaps lack of chip mem on A500 compared to A1200? Shame because it was looking both unbelievable and like it was going to work.

Cego · 25 February 2018, 23:10

Do you think the original PSU (18Amps on 5V and 4Amps on 12V) is powerful enough to supply following config?

- A4000D with 16MB onboard Ram
- Sonnet G4 @400MHz + 192MB Ram
- CS MKII with SCSI Kit
- Yamaha SCSI IDE Adaptor + 80GB mSata SSD
- Mediator 4000Di
- Voodoo 3 2000
- Realtek Ethernet card
- DVD drive (3,5“, bulk)
- 1x Floppy drive

Cego · 25 February 2018, 23:10

if not, which PSU do you recommend?

Cego · 26 February 2018, 01:02

according to a user on a1k.org his A4000 is consuming 58Watts with CS MKII-060, CV64, A2320, FASTATA, DVD-Burner and two 2,5HDs.

Sonnet G4 probably eats about 30Watts and the Voodoo 3 2000 about 20Watt. That would sum up to 110-120 Watts which is still in the parameters (145Watt) of the original PSU. This should be fine right?

nexus · 26 February 2018, 01:15

maybe also depends on how healthy the psu is they are old and now being ran at there limit

when capacitors start dying they cant take power loads very well

hedeon still hasnt ran 1 due to same issue bad caps he has to let it warm up some to get anywhere which means the caps are having issues holding a charge which = going bad or has failed and another cap is taking on the stress which then makes it fail faster

grelbfarlk · 26 February 2018, 01:17

Quote:

Originally Posted by Cego

according to a user on a1k.org his A4000 is consuming 58Watts with CS MKII-060, CV64, A2320, FASTATA, DVD-Burner and two 2,5HDs.

Sonnet G4 probably eats about 30Watts and the Voodoo 3 2000 about 20Watt. That would sum up to 110-120 Watts which is still in the parameters (145Watt) of the original PSU. This should be fine right?

I don't think the Sonnet G4 is at 30W, I think it's more like 10-20W.
http://eab.abime.net/showpost.php?p=...postcount=1523

From that post:
Heavily loaded tower running Q2 Timedemo=104 Watt.

Now that is Rapture not Sonnet and it has the 7410 which is a bit more power efficient than the 7400 of the Sonnet G4.
Worst case I'd guess an A4000 with Sonnet G4 should be not be much more than 90W, which is pretty high and most of that load is on your 5V rail, not all of it of course. Generally speaking you'd not want to run at 100% of your max 5V load on any power supply, definitely not a 25 year old one.

So yeah it's probably going to be ok but it's in the edgy territory.

Also consider the talk pages back about the 3A VRM of the Mediator Black edition. On the Sonnet the RAM power supply is on 5V and the rest of the board is 3.3V. The specs state that a 7400@400MHz consumes a maximum of 11.3W. That alone is over the rating of the 3.3V@3A of the VRM on the Mediator. So you might consider modifying the Sonnet with its own VRM. Since the rest of the board, L2 cache, MPC107 and the other VRMs on the board for 2.5V, 2.2V are going to consume something as well.

Cego · 26 February 2018, 01:50

thanks for the info! is there a guide on how to modify the Sonnet?

nexus · 26 February 2018, 02:07

on the wiki page

also i think its time to update that profile pic

Cego · 26 February 2018, 02:16

this one?
https://github.com/Sakura-IT/SonnetA...0-power-supply

so by adding the additional voltage regulator the sonnet will have sufficient 3,3V power supply?

17 February 2018, 21:59	#1746
Wrangler Registered User Join Date: Sep 2015 Location: London, UK Posts: 419	A500+/ACA+/Mediator Sadly, no amount of fiddling around with 1200TX mem window jumpers, MMU envs (and all the others), ACA500+ parameters etc makes any difference. I think this is a simple case of a step too far with hardware that was never designed to work together.

18 February 2018, 12:24	#1747
nexus Registered User Join Date: Aug 2017 Location: USA Posts: 728	do you have latest firmware for the aca500+? it uses a new MMU library http://wiki.icomp.de/wiki/ACA500plus might need to play with those settings also http://wiki.icomp.de/wiki/ACA500plus_registers Software-defined autoconfig This feature helps getting drivers into the Amiga OS without the need for ugly hacks. With some A1200 accelerators "covering" the $f0.0000 area, the old method that was used on the ACA500 had to be improved. The solution is to sacrifice 64k of memory (of which "plenty" is available) and prepare it with autoconfig data. Although any vendor ID and product ID can be used, you should use the iComp vendor ID 4626 ($1212) and product ID 65 ($41) for the autoconfig nibbles. These have already been added to the latest boards.library. Since there is no config_in and config_out lines on the A1200 accelerator port, the ACA500plus implements a "sniffing circuit" that will generate these signals internally. For this circuit to work properly and maintain the config chain, the system must know ahead of time which address the 64k-sized board will be mapped to, and how many Autoconfig boards are located within the A1200 accelerator. A few examples: ACA12xx accelerators use no autoconfig at all. The ACA500plus should be programmed to be the first autoconfig board. ACA1233n accelerators use Z3 autoconfig for it's memory, which will end up outside the Z2 area. The ACA500plus should be programmed to be the second autoconfig board and to final address $e9.0000. ACA1221 accelerators use Z2 autoconfig for a 64k-sized IO space, which will be mapped to $e9.0000. The ACA500plus should be programmed to be the second autoconfig board and to final address $ea.0000. Blizzard 1230-IV accelerators use Z2 autoconfig for a 128k-sized IO space, which will be mapped to $ea.0000. The ACA500plus should be programmed to be the second autoconfig board and to final address $e9.0000. Last edited by nexus; 18 February 2018 at 12:34.

26 February 2018, 02:07	#1759
nexus Registered User Join Date: Aug 2017 Location: USA Posts: 728	on the wiki page also i think its time to update that profile pic Last edited by nexus; 26 February 2018 at 02:19.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
WANTED: PCI Sound Card VORTEX2 SuperQuard / SQ2500 or Turtlebeach SantaCruz pci card	keropi	MarketPlace	12	15 March 2017 02:48
Wanted:Sonnet Technologies QuadDoubler 50 (Doubler 4000)	kronuz	MarketPlace	0	16 June 2013 00:29
eBay : Sonnet Quaddoubler	alexh	MarketPlace	3	07 May 2013 00:50
Overlay problem on GeForce 7200	Leandro Jardim	support.WinUAE	1	10 August 2010 15:50
Prometheus PCI & Voodoo 3 PCI GFX Card	Slayer	support.Hardware	21	05 September 2006 10:57

16 February 2018, 15:42	#1742
Wrangler Registered User Join Date: Sep 2015 Location: London, UK Posts: 419	Update on the Frankenfest (A500+/ACA500+/Mediator): the mediator autoconfigs, PCIINFO recognises the Voodoo card but any attempt to open the Voodoo monitor driver causes a guru of various different types. It seems close to working but my suspicion is that there is a memory problem - perhaps lack of chip mem on A500 compared to A1200? Shame because it was looking both unbelievable and like it was going to work.

16 February 2018, 17:09	#1743
trixster Guru Meditating Join Date: Jun 2014 Location: England Posts: 2,354	I got gurus when opening the voodoo monitor driver when I didn't have layers.library v45 installed correctly. Are you sure p96 etc is setup correctly?

16 February 2018, 18:33	#1744
Wrangler Registered User Join Date: Sep 2015 Location: London, UK Posts: 419	I'm pretty sure the Picasso install is good and I reinstalled it for good measure with the same results. I'm running this with an '030 - that wouldn't make any difference?

16 February 2018, 23:14	#1745
grelbfarlk Registered User Join Date: Dec 2015 Location: USA Posts: 2,958	'030 should be fine but if you're using A1200 Mediator you might need to play with the MMU variable if it exists or not. P96 I think requires a minimum of 020. Also make sure you're not using all of the 8MB of Zorro-II space maybe?

22 February 2018, 01:13	#1748
Joe Registered User Join Date: Oct 2017 Location: Slovakia Posts: 42	Hello everyone, I would like to inform about my sonnet issue: - My sonnet was tested by Jack3D and working - My RAM which I borrowed to Jack3D one piece suddenly began working on last time when he tested - Jack3D sent back to me sonnet and I tested with new RAMs which I ordered and with the same result - Ramlib Error after inicalizing powerpc.library - I bought Voodoo 3 3000 and I replaced the gfx card. I booted Jack3D system and after initializing powerpc.library the same error was occured - RamLib Error So, already I see the RamLib Error if I use Radeon 9200 gfx or Voodoo 3 3000 gfx. So, I do not know where the problem might be. Jack3D told me that I have strange mediator. My mediator working fine without sonnet or with sonnet but without inicializating initppc command. My mediator was produced in 2001 year. I see that working correctly. What do you thing where can be a problem? BTW what does it mean this error - RamLib Error 80000004 , I tried some patch tools as PatchRam, MCPRAMLibPatch or etc. but without successfull. And one more thing. What error will be showed when you added bad RAM in your cause. What error was showed like in my cause or other error? Thank you very much for your tips.

24 February 2018, 00:19	#1749
grelbfarlk Registered User Join Date: Dec 2015 Location: USA Posts: 2,958	I'm not understanding a little bit. You had a stick of RAM which didn't work for you but suddenly started working for Jack. The same board with the same RAM didn't work when you got it back? But with totally different RAM which should work, it doesn't. Like I was saying in the private messages, usually RamLib 80000004 is a 68k crash. So what else is different between your system an Jack's? Is he using a custom 3.9 ROM and you are using original KS3.1? Also as I said in the PMs here are some of the library error messages which have been associated with but not necessarily caused by memory issues: "No memory detected on the Sonnet card" "Memory corruption detected during setup" "PowerPC CPU possibly crashed during setup" "PowerPC CPU not responding" Also do all memory slots work the same? Depending on revision on the CPU side of the board there should be small SMT resistors that were added later to the back of the memory slots, I've had two boards which one of the resistor legs were not soldered just right. Caused sometimes memory to show up or crash when Initing. If you are able, very gently try to check the resistance of them on the pins of the socket and maybe very very gently try to wiggle it slightly to see if the connection is broken. Don't know if you did this but if you have the one old DIMM working try just that one first. Then replace with just one new DIMM.

24 February 2018, 01:50	#1750
jack-3d kLiker Join Date: Mar 2011 Location: Brno / Czech Republic Posts: 371	Simply he sent me his sonnet and ram. I checked and it was not working. I replaced ram by my module. It was working fine. Last day just before packing I tested his ram again, just to make screenshot of the error and suddenly his ram was working (even after few restarts). Miracle? Anyway I send him back card, ram and also disk with my system fully tested. But he still has the ram lib error. So... Sonnet = tested, working RAM = tested, one day not working, next day working OS = tested, working Voodoo = not tested mediator = not tested mainboard = not tested I would bet on mediator.

24 February 2018, 02:36	#1751
nexus Registered User Join Date: Aug 2017 Location: USA Posts: 728	sounds more like a weak solder joint on the sonnet ram slots not sure how you guys are setting them up but its easy to flex the sonnet while inserting and if ram is already installed then it could be rocked out of the slot so after installing make sure its re-seated and clean fyi you shouldnt mix ram that use different plating on its contacts eg gold and tin if the slots are tin use tin plated ram sticks vice versa for gold

25 February 2018, 23:10	#1753
Cego Registered User Join Date: Dec 2015 Location: Germany Posts: 163	Do you think the original PSU (18Amps on 5V and 4Amps on 12V) is powerful enough to supply following config? - A4000D with 16MB onboard Ram - Sonnet G4 @400MHz + 192MB Ram - CS MKII with SCSI Kit - Yamaha SCSI IDE Adaptor + 80GB mSata SSD - Mediator 4000Di - Voodoo 3 2000 - Realtek Ethernet card - DVD drive (3,5“, bulk) - 1x Floppy drive

25 February 2018, 23:10	#1754
Cego Registered User Join Date: Dec 2015 Location: Germany Posts: 163	if not, which PSU do you recommend?

26 February 2018, 01:02	#1755
Cego Registered User Join Date: Dec 2015 Location: Germany Posts: 163	according to a user on a1k.org his A4000 is consuming 58Watts with CS MKII-060, CV64, A2320, FASTATA, DVD-Burner and two 2,5HDs. Sonnet G4 probably eats about 30Watts and the Voodoo 3 2000 about 20Watt. That would sum up to 110-120 Watts which is still in the parameters (145Watt) of the original PSU. This should be fine right?

26 February 2018, 01:15	#1756
nexus Registered User Join Date: Aug 2017 Location: USA Posts: 728	maybe also depends on how healthy the psu is they are old and now being ran at there limit when capacitors start dying they cant take power loads very well hedeon still hasnt ran 1 due to same issue bad caps he has to let it warm up some to get anywhere which means the caps are having issues holding a charge which = going bad or has failed and another cap is taking on the stress which then makes it fail faster

26 February 2018, 01:50	#1758
Cego Registered User Join Date: Dec 2015 Location: Germany Posts: 163	thanks for the info! is there a guide on how to modify the Sonnet?

26 February 2018, 02:16	#1760
Cego Registered User Join Date: Dec 2015 Location: Germany Posts: 163	this one? https://github.com/Sakura-IT/SonnetA...0-power-supply so by adding the additional voltage regulator the sonnet will have sufficient 3,3V power supply?

Currently Active Users Viewing This Thread: 2 (0 members and 2 guests)