Time to get off Firewire? Where to?

Configure and optimize you computer for Audio.
RELATED
PRODUCTS

Post

mutools wrote:
pljones wrote:2) Swap the application API -- MuLab works with the MME for audio output to the hardware -- that also swaps out the ASIO driver, of course, as they "bound" together.
Are you sure about that? Maybe the failure is in the Echo ASIO specific code of the Echo driver. (outside the MME specific code)
Yes, that's what this test is about - the ASIO code in the Echo driver. This test bypasses that code. It shows the problem is related to the Echo ASIO code. However that's the same code used by all other ASIO applications and they work and haven't changed, so it's not a general failure. It also worked with earlier versions of MuLab, and the ASIO code hasn't changed. So it does seem to be specific to later versions of MuLab and the unchanged Echo code.
mutools wrote:
3) Swap the ASIO driver -- MuLab works with ASIO out to, e.g. Reaper via ReaRoute, with Reaper's main out to the Echo.
You could also try Asio4All.
I'd expect it to work as it doesn't sit on the Echo ASIO driver at all but it would test another part of the Echo code, true (i.e. the DirectAudio or whatever it is). I'd expect it to work even if it ran with the ASIO driver, as most applications still do. Even if ASIO4ALL had an issue, that would not necessarily mean it was related to the problem here - although it might be interesting to confirm whether it acted the same for all its clients. I may try it next weekend.
mutools wrote:And FYI: I did not receive any other similar report wrt M7.4 about failing ASIO functionality.
Ok, that doesn't 'prove' anything, but at least it's relevant statistical info.
Yes, agreed - it's definitely not a general issue. That does make it more frustrating :).
mutools wrote:
To me it looks like it's the interaction between the application and the ASIO driver. One side or other isn't "quite right" and - for the specific combination of MuLab and the Echo ASIO drive - this causes a problem.
Not only MuLab, the ASIO SDK hostsample doesn't work either. (when also built with VS2015/2017)
Sorry, yes, I was not meaning to imply otherwise - just summarising the position to that point.
mutools wrote:
Note that compiling hostsample with VS2013 or VS2017 produced the same result -- no output -- so it doesn't appear to be compiler version. It could be some (Microsoft-updated) compile or runtime library.
You also built the ASIO SDK hostsample with VS2013 and it's also not working with the Echo Firewire? (didn't knew that, sorry for double-checking the VS2013 aspect)
Yes, 2013 and 2017 -- I don't have 2015 and it had become unavailable.
mutools wrote:And as firewire seems to be dying and as they're not updating drivers anymore, what about switching to another audio interface? Is that an option for you? (pragmatic solution)
Yes, that was my thinking in starting the thread. I really do see Firewire dying as an option in the near future. The problem is, MIDI seems to be almost an after-thought to most audio developers - zero connectors or a dongle/breakout cable.

(The breakout cables usually take up significant amounts of space compared with that needed for not having such a cable -- that's why they're not an option, for those interested. Things would start falling damaging distances...)

Post

pljones wrote:The problem is, MIDI seems to be almost an after-thought to most audio developers - zero connectors or a dongle/breakout cable.
My recommendations to you, in order, are the following:
https://www.roland.com/us/products/quad-capture/
https://www.native-instruments.com/en/p ... e-audio-6/
http://www.presonus.com/products/AudioBox-USB-96
https://www.steinberg.net/en/products/a ... 2mkii.html

I owned and used the first two, my brother owns the third, and the last one seems to be well regarded overall. I believe all meet your requirements from the first post. I still use the Roland and love it.
Feel free to call me Brian.

Post

Hi bmrzycki -- yes, they do! Thanks. The Roland looks pretty much like it would do perfectly. Not keen on the look of the NI box (just looks!! :) ) Can't seem to find the Presonus for sale in the UK (I think I'm seeing an older model in the ads), otherwise that looked good, too. the UR22mk2 also looks good, and a great price on that one for what you get, too! I'll have a hunt around KVR for anything on the four but at least that's the form-factor I'm after, which saves me a huge amount of work. Thanks again!

Post

I really like the Roland Quad Capture. They are the first USB-powered audio interface that never generates pops, clicks, noise when moving the volume knob. They did this by physically isolating the power conversion to a separate board inside the chassis. That combined with a ground lift switch also means I haven't experienced any ground loops or hum issues: something that plagued other devices I tried before. My only gripe (a small one) is there is no separate channel or volume knob for headphones. I have my monitors on a separate power strip so it's easy for me to turn them off when I just want to use headphones, but others may be annoyed.
Feel free to call me Brian.

Post

pljones wrote:
mutools wrote:
pljones wrote:2) Swap the application API -- MuLab works with the MME for audio output to the hardware -- that also swaps out the ASIO driver, of course, as they "bound" together.
Are you sure about that? Maybe the failure is in the Echo ASIO specific code of the Echo driver. (outside the MME specific code)
Yes, that's what this test is about - the ASIO code in the Echo driver. This test bypasses that code. It shows the problem is related to the Echo ASIO code. However that's the same code used by all other ASIO applications and they work and haven't changed, so it's not a general failure. It also worked with earlier versions of MuLab, and the ASIO code hasn't changed. So it does seem to be specific to later versions of MuLab and the unchanged Echo code.
As we both know MuLab 7.3.9 was working fine with the Echo Firewire. Now i'm 99.99% sure that if i simply recompile M7.3.9 with VS2015 then the issue will be there ie no audio output, just like with the standard Steinberg ASIO SDK hostsample. So the Mu code has not changed at all but the issue arrives. The only thing that changed was the Windows libraries. Maybe there is an incompatibility between the (older) Echo ASIO driver and newer Windows libraries? It's all guessing, the Echo dev team is the one who should step in here. But i understood there not really interested, right?
mutools wrote:And as firewire seems to be dying and as they're not updating drivers anymore, what about switching to another audio interface? Is that an option for you? (pragmatic solution)
Yes, that was my thinking in starting the thread. I really do see Firewire dying as an option in the near future. The problem is, MIDI seems to be almost an after-thought to most audio developers - zero connectors or a dongle/breakout cable.

(The breakout cables usually take up significant amounts of space compared with that needed for not having such a cable -- that's why they're not an option, for those interested. Things would start falling damaging distances...)
Like bmrzycki, i can also recommend Roland. I use a Duo Capture Ex and am very happy about it. Solid hardware and software, built-in MIDI.

Post

I built a new rig 2 years ago, Xeon, ECC ram, all the rest.
My problems started with the firewire card itself (good luck finding a FW400 built into the motherboard nowadays) - 1 wasn't so great, the other was better, but still huge problems (this is under Windows 7 - I wouldn't touch windows 10 with a 10 foot pole). Eventually I discovered the onboard graphics card caused too many spike problems, so I got an old nvidia card - which was lucky because the newer nvidia drivers also cause spikes with firewire.
Luckily this one was supported by a particular driver version set (185.85) which doesn't cause the spikes.
Don't bother believing the spike tools as to what's causing it - they're all wrong.
Now, it's pretty good. Not as good as it was on XP (at least in terms of firewire) but good enough.
Running a MOTU 896HD upgraded by BLA chained to a ultralite.
Good luck, but I would go with USB3 if you can.

Post

What board was in that Xeon rig? What FW400 cards were you using?

The TI FW400 cards haven't really worked all that well for a few generations of motherboard now and the TI controller chip on the FW400 cards has been legacy for almost a decade now, with support long having dropped off. Entirely possible that the newer Nvidia drivers were conflicting with the no longer supported chipset.

FW800 cards tend to work fine still, if it was still failing to run smoothly with one of those, ignore the above. I also now find myself wondering how applicable this is to VIA and the rest of them too!

Post

Kaine, what's your verdict on the Agere-powered FW cards? As they're the only native PCIe option that's what I would go for if I was looking to add firewire. But then I'm reading reports from people who are using TI-powered cards without any issues. Is that because they're using FW800 TI chips or is it just pot luck?

I looked into all this a few years ago when I was putting together my X79 rig and concluded that firewire was on the way out and it was better to just stick with USB.
Image Image Image Image

Post

sprnva wrote:Kaine, what's your verdict on the Agere-powered FW cards? As they're the only native PCIe option that's what I would go for if I was looking to add firewire. But then I'm reading reports from people who are using TI-powered cards without any issues. Is that because they're using FW800 TI chips or is it just pot luck?

I looked into all this a few years ago when I was putting together my X79 rig and concluded that firewire was on the way out and it was better to just stick with USB.
TI Chipset FW400, ATI HD 5570 video card, Intel Z97 chipset (4790K CPU), RME Fireface 400.

This is the system I'm going to lock down and use (with Windows 7) for the next 5+ years, as it's been flawless since I put it together.
Remember the iLokalypse Summer 2013

Samples and presets and free stuff!

Post

sprnva wrote:Kaine, what's your verdict on the Agere-powered FW cards? As they're the only native PCIe option that's what I would go for if I was looking to add firewire. But then I'm reading reports from people who are using TI-powered cards without any issues. Is that because they're using FW800 TI chips or is it just pot luck?
It's because they are using the FW800 variant.

Agere works fine for some interfaces so I hear, although personally I don't use them simply because I source and have easy access to the TI's so I've not gone and tested any other models to be able to comment further.
Dominus wrote: TI Chipset FW400, ATI HD 5570 video card, Intel Z97 chipset (4790K CPU), RME Fireface 400.

This is the system I'm going to lock down and use (with Windows 7) for the next 5+ years, as it's been flawless since I put it together.

You were lucky to get any TI XIO2200A based card working at all on that chipset, I know we saw few working combinations and they certainly don't on the generations since.

It's interesting to note that even TI don't recommend using that chip now and haven't for some time.

http://www.ti.com/product/XIO2200A

The suggested replacement that they list there is the controller that we've been using in house for a while and everything I've tested (so far at least) has behaved well on it.

Post

Kaine wrote:
You were lucky to get any TI XIO2200A based card working at all on that chipset, I know we saw few working combinations and they certainly don't on the generations since.

It's interesting to note that even TI don't recommend using that chip now and haven't for some time.

http://www.ti.com/product/XIO2200A

The suggested replacement that they list there is the controller that we've been using in house for a while and everything I've tested (so far at least) has behaved well on it.
Is that the only chip they used? This is an older PCIe firewire card, it's been through a number of systems.
Remember the iLokalypse Summer 2013

Samples and presets and free stuff!

Post

Dominus wrote: Is that the only chip they used? This is an older PCIe firewire card, it's been through a number of systems.
AFAIK the two revisions of that chip were it for purely FW400 cards although I may have missed one. The XIO2200A/B chips were in use for well over a decade and recommended by everyone as the best working solution for A/V kit, then around your board generation it seemed like the majority of boards just didn't like them and by Z170 / X99 after yours as far as I've seen first hand they just don't work full stop.

Post

Kaine wrote: AFAIK the two revisions of that chip were it for purely FW400 cards although I may have missed one. The XIO2200A/B chips were in use for well over a decade and recommended by everyone as the best working solution for A/V kit, then around your board generation it seemed like the majority of boards just didn't like them and by Z170 / X99 after yours as far as I've seen first hand they just don't work full stop.
Never had a problem with it. The replacement is $21 on Amazon if I was to ever upgrade, so I'm not worried. I'll give up my Fireface when RME stops supporting it. :D
Remember the iLokalypse Summer 2013

Samples and presets and free stuff!

Post

Kaine wrote:What board was in that Xeon rig? What FW400 cards were you using?
Can't remember, will try to get it when next at studio.
Kaine wrote:Entirely possible that the newer Nvidia drivers were conflicting with the no longer supported chipset.
No, it's an established problem with any nvidia drivers after the mentioned version. Occurs with newer nvidia cards as well as older ones, I tested at the time. ATI worked fine. I wrote about it on gearslutz at the time (actually references a post you made - I think?):
https://www.gearslutz.com/board/11273725-post14.html

There's a bunch of other stuff you've got to do to get it working properly as well, all listed there.

Post

Kaine wrote:The legacy Texas Instruments chip that no longer works is the TI XIO 2200A.

The current one that is known to be good is the TI XIO 2213B and is found on the triple port cards normally in a 2 X 800 + 1 X 400.
For what it's worth, I bought an Iocrest card (as it says on the box -- "Syba" as it's called on Amazon; same product number) with that chipset, hoping to use it with a Saffire Pro 40 I just bought. No luck at all -- Win 7 Device Manager has it failing with code 10. I tried it in all of my PCIe slots, even moving my video card. No difference.

The legacy driver doesn't have an error message, but doesn't show up either in OHCI Tool or in Saffire MixControl.

My motherboard (about 6 years old, MSI X58 Pro-E) has a FireWire 400 port. I ordered a cable to see if I can get that working, but OHCI Tool doesn't like it so it's not likely.

Any ideas? Did I just get a dud interface card, maybe?

Post Reply

Return to “Computer Setup and System Configuration”