Ricoh cadbus controllers
F**k! I bought a really great laptop the other day, thinking of getting a Magma/IBP expansion system, but I just discovered the alleged problems with Ricoh cardbuses, which sadly my laptop has.
I guess there is no way to work around this problem. Maybe I need to part with my Creamware cards now... :/
I guess there is no way to work around this problem. Maybe I need to part with my Creamware cards now... :/
I have actually ordered that sollution now, so my setup will either come to a complete halt or perform wonderfully. Having been in contact with Ricoh I see no reason why their controllers should perform any less.On 2005-02-26 17:22, stardust wrote:
maybe there are other solutions than magma ?
http://www.planetz.com/forums/viewtopic ... 4&forum=10
Another thing, both Magma and IBP have removed their laptop guides it seems, with system specs for 2003-systems and less.. Magma still warn about the Ricoh-controllers, though IBP do not, and IBP are probably OEM Magmas.
There are some warnings against Ricoh-controllers on RME's site too, but their reports are likewise from old systems, though they actually reported a working system, based on PIII and propertiary chipsets.
I have a completely up to date Centrino system here, sporting an up-to-date Ricoh CardBus-controller, which according to Ricoh should perform PCI Bursts, theoretically allowing for a bandwidth up to 132MB/s.
So, I will be testing this and I will of course post my findings here. Either it works, or either I will have to part with the laptop or the IBP/Creamware gear..
<font size=-1>[ This Message was edited by: voidar on 2005-02-26 20:03 ]</font>
Hehe..
---------------------
Answer]
We are pleased to receive your inquiry about our device product.
> Is the Super One Chip R5C592 a proper 32-bit Cardbus controller?
Yes, R5C592 is a proper 32-bit Cardbus controller.
> So, another question would be; is the S.O.C. R5C592 really capable of
> +100 MB/s peak performances (PCI Bursts), or is this just a R5C476II
> controller issue?
Our R5C592 has a capability of +100MB/sec at peak performances (PCI Bursts).
-----------------------------
I actually asked about the wrong controller. I have the R5C593, but I guess it has the same CardBus functions. The difference is that it supports more media cards.
But who really knows....
---------------------
[Answer]
We are pleased to receive your inquiry about our device product.
> Is there any real difference between the 5C476(II) and the CardBus
> Controller of both R5C592 and R5C593?
Although R5C476(II), R5C592 and R5C593 have the same Device ID,
they have some differences in thier CardBus Controller functions.
Sorry, we can't explain the differences in detail on this system,
Please contact our sales representative.
The following is thier information.
Company name : Unique MEMEC Sweden
Contcat person & E-mail Address :
"Bianca Pemberton" bpemberton@unique.eu.memec.com
"Patrik Sarenfors" psarenfors@unique.se.memec.com:
Sincerely Yours,
------------------
Anyway, the old tests were done using R5C476(II). I am just crossing my fingers for that this will work.
<font size=-1>[ This Message was edited by: voidar on 2005-02-27 12:35 ]</font>
---------------------
Answer]
We are pleased to receive your inquiry about our device product.
> Is the Super One Chip R5C592 a proper 32-bit Cardbus controller?
Yes, R5C592 is a proper 32-bit Cardbus controller.
> So, another question would be; is the S.O.C. R5C592 really capable of
> +100 MB/s peak performances (PCI Bursts), or is this just a R5C476II
> controller issue?
Our R5C592 has a capability of +100MB/sec at peak performances (PCI Bursts).
-----------------------------
I actually asked about the wrong controller. I have the R5C593, but I guess it has the same CardBus functions. The difference is that it supports more media cards.
But who really knows....
---------------------
[Answer]
We are pleased to receive your inquiry about our device product.
> Is there any real difference between the 5C476(II) and the CardBus
> Controller of both R5C592 and R5C593?
Although R5C476(II), R5C592 and R5C593 have the same Device ID,
they have some differences in thier CardBus Controller functions.
Sorry, we can't explain the differences in detail on this system,
Please contact our sales representative.
The following is thier information.
Company name : Unique MEMEC Sweden
Contcat person & E-mail Address :
"Bianca Pemberton" bpemberton@unique.eu.memec.com
"Patrik Sarenfors" psarenfors@unique.se.memec.com:
Sincerely Yours,
------------------
Anyway, the old tests were done using R5C476(II). I am just crossing my fingers for that this will work.
<font size=-1>[ This Message was edited by: voidar on 2005-02-27 12:35 ]</font>
Got it today. Really, it's larger than I thought. A bit hard to connect with such a short cable.
Anyway, I got my two pulsars in there. I seem to be able to load about half of the DSPs with low-bandwidth plugs like Vinco etc. It craped-out at once loading a Masterverb. I mean, the system freezes. I managed to get the "PCI-bandwidth..." error when loading one of my old mix-projects.
This is really sad. Maybe I'll migrate to SAW, or I could always sum outside the box with some nice Firewire interfaces, that is if the cardbus-controller can mange a firewire interface..
Or perhaps I should go barebone.
Anyway, I am selling the CardBox.
Anyway, I got my two pulsars in there. I seem to be able to load about half of the DSPs with low-bandwidth plugs like Vinco etc. It craped-out at once loading a Masterverb. I mean, the system freezes. I managed to get the "PCI-bandwidth..." error when loading one of my old mix-projects.
This is really sad. Maybe I'll migrate to SAW, or I could always sum outside the box with some nice Firewire interfaces, that is if the cardbus-controller can mange a firewire interface..
Or perhaps I should go barebone.
Anyway, I am selling the CardBox.
I was once looking into the rack-mount option. I figured it just wasn't as portable as I wanted it really. But I really don't think Magma/IBP gear is anyway, now.On 2005-03-03 12:05, notbobmoog wrote:
A very fast & robust 3U rackmounted PC, Iu keyboard shelf & a Noah will fit nicely into a 6U SKB or similar (i meant cheaper).
Not as portable, not as flexible, but probably cheaper & more reliable/robust.
+++ you could get 3 CW cards in there !!
Nice![]()
I sort of like this laptop. I got an external firewire LaCie HD for too. I guess I have to do some more testing.
I have no idea if it's doable or not, but what about assigning a much higher pci latency to the cardbus controller?
This is a freeware I'm using with no problem. it has been suggested here before..
http://194.71.11.70/pub/games/PC/guru3d ... u3d%5D.zip
This is a freeware I'm using with no problem. it has been suggested here before..
http://194.71.11.70/pub/games/PC/guru3d ... u3d%5D.zip
What is your setup? You are using this to decrease the possibility of PCI bandwidth errors?On 2005-03-03 16:57, alfonso wrote:
I have no idea if it's doable or not, but what about assigning a much higher pci latency to the cardbus controller?
This is a freeware I'm using with no problem. it has been suggested here before..
http://194.71.11.70/pub/games/PC/guru3d ... u3d%5D.zip
I used to check pci latency values on my music pci that is fine and didn't need tweak, but I have also my internet pci with a luna card and an ATI gr card, and I used to reduce ATI's very high default value, stopping the stuttering of audio playbacks....
Normally, higher the value, more resources are assigned, so you might try to raise cardbus controller pci latency and lower graphics....if it can be done, as I don't have a laptop and can't say for sure.
But it's a non installation program, very light and didn't harm my computer at all...
Normally, higher the value, more resources are assigned, so you might try to raise cardbus controller pci latency and lower graphics....if it can be done, as I don't have a laptop and can't say for sure.
But it's a non installation program, very light and didn't harm my computer at all...
I tried this, though I never got it passed 255. It was set at 248.On 2005-03-03 17:18, alfonso wrote:
I used to check pci latency values on my music pci that is fine and didn't need tweak, but I have also my internet pci with a luna card and an ATI gr card, and I used to reduce ATI's very high default value, stopping the stuttering of audio playbacks....
Normally, higher the value, more resources are assigned, so you might try to raise cardbus controller pci latency and lower graphics....if it can be done, as I don't have a laptop and can't say for sure.
But it's a non installation program, very light and didn't harm my computer at all...
I installed a pci to pci bridge driver this time though... At least the system did not freeze as easily, though I could not load much this time.
I think I am going firewire..
<font size=-1>[ This Message was edited by: voidar on 2005-03-03 18:32 ]</font>
I've been using Magma for a while (with TI controllers on Dell laptops). I have *always* had the PCI-PCI Bridge installed - as far as I know it's impossible to use the Magma properly without it.
Also, try *decreasing* PCI latency of some other devices in your system, especially the graphics card.
And yes, generally on laptops the PCMCIA cardbus shares IRQ with almost everything else, so try and use as little as possible of the other stuff. It's frustrating I know... things like this make me get closer and closer to going Mac..
<font size=-1>[ This Message was edited by: dArKr3zIn on 2005-03-24 11:45 ]</font>
Also, try *decreasing* PCI latency of some other devices in your system, especially the graphics card.
And yes, generally on laptops the PCMCIA cardbus shares IRQ with almost everything else, so try and use as little as possible of the other stuff. It's frustrating I know... things like this make me get closer and closer to going Mac..

<font size=-1>[ This Message was edited by: dArKr3zIn on 2005-03-24 11:45 ]</font>
Hmm, I have been disabling a lot of stuff I don't need on my lappy. I might have another go at this actually.On 2005-03-24 11:44, dArKr3zIn wrote:
I've been using Magma for a while (with TI controllers on Dell laptops). I have *always* had the PCI-PCI Bridge installed - as far as I know it's impossible to use the Magma properly without it.
Also, try *decreasing* PCI latency of some other devices in your system, especially the graphics card.
And yes, generally on laptops the PCMCIA cardbus shares IRQ with almost everything else, so try and use as little as possible of the other stuff. It's frustrating I know... things like this make me get closer and closer to going Mac..
<font size=-1>[ This Message was edited by: dArKr3zIn on 2005-03-24 11:45 ]</font>
First time I did not install the PCI-PCI driver, but I did so the second time. The differences was that the system didn't reboot, rather freeze instead.
Is the Magma driver and the PCI-PCI driver the same thing or?
My system is really strange really. I have USB host-controllers and firewire-controllers without physical interfaces.
System looks like this (not listed ISA-connections):
3 ATI Mob. Rad. 9700
3 USB host-cont.
9 OHCI Compliant IEEE 1394..
9 Realtek AC'96 Audio
9 Realtek ethernet
10 USB host-cont.
10 Ricoh 553 s000pad000pachip
11 Ricoh 553 s000paD000pachip
I have an external firewire disc now, and I bought one of these CardBus-firewire adapters.
I could actually get 40MB/s bursts from both my firewire interfaces measured using HD Tach.
Would be interesting to test FW800 cabability, to really see what the Ricoh controllers can do. Often I am tempted into believing much of the blame is on Magma too for having one-dimentional drivers.
Something looks very strange there... probably best to reinstall the OS. The PCI-PCI Bridge has always been present since I've been using the Magma.
One thing which is extremely important is to follow the Magma driver installation *EXACTLY*. I've never had any problems with them at all, except when I tried to install it on a very sub-standard Sony machine with a ricoh cardbus controller. I'm still not sure if it was the cardbus or the motherboard of the Sony itself - I personally think that the motherboard is much more crucial with this stuff.
One thing which is extremely important is to follow the Magma driver installation *EXACTLY*. I've never had any problems with them at all, except when I tried to install it on a very sub-standard Sony machine with a ricoh cardbus controller. I'm still not sure if it was the cardbus or the motherboard of the Sony itself - I personally think that the motherboard is much more crucial with this stuff.
The lap is going out the window this tuesday. I am getting a new one which has Texas Instrument chips that are actually said to be supported by Magma. a Clevo M400A.On 2005-03-27 17:54, dArKr3zIn wrote:
Something looks very strange there... probably best to reinstall the OS. The PCI-PCI Bridge has always been present since I've been using the Magma.
One thing which is extremely important is to follow the Magma driver installation *EXACTLY*. I've never had any problems with them at all, except when I tried to install it on a very sub-standard Sony machine with a ricoh cardbus controller. I'm still not sure if it was the cardbus or the motherboard of the Sony itself - I personally think that the motherboard is much more crucial with this stuff.
http://www.clevo.com.tw/products/M400A.asp
It is identical to the Sager 4880. http://www.sagernotebook.com/pages/note ... tType=4880
Actuallt, Sager, Eurocom, Alienware, Voodoo
PC etc. are Clevo OEM, similar to how IBP is a Magma OEM product.
Anyway, it will be nice to be able to use my Creamware stuff again, as I refuse to lose any more money. I will get a lousy 956€ of this laptop and it will pay most of the new one. The Clevo is actually a Sonoma system that uses DDR1 memory, which is ok by me as I will get to use the 1GB from my current one. Also, I get a GPU upgrade to an ATI x700.
I would have jumped on the new Dell XPS2, but sadly it has Ricoh chips..
There is no coinsidence that Magma warn against the Ricoh chips, so have/do RME. And even some other vendors.
So, stay away from MSI, Asus, Acer etc. as you do not know what you will get. Clevo/Sager seems to be an ok brand.