Gotta be the quote of the day.peakdesign wrote:I think you can either be realistic and work with the stuff, understand shortcomings, and work around limitations, or you gonna be feeling road rage while chasing rainbows.
CME UF8 USB/MIDI Master Controller Keyboard - USB Issues
-
- KVRAF
- 2427 posts since 16 Jun, 2005 from Somewhere, NV
-
- KVRAF
- 6323 posts since 30 Dec, 2004 from London uk
bobsled wrote:peakdesign wrote:I think you can either be realistic and work with the stuff, understand shortcomings, and work around limitations, or you gonna be feeling road rage while chasing rainbows.
Such product loyalty is amazing. I just use what works and ditch the rest.
-
- KVRer
- Topic Starter
- 21 posts since 3 Jul, 2004 from Staffordshire
Folks,
[OK, I've actually got a 'free' lunch-break, so time to rant - Ed!]
There's absolutely no criticism of anybody involved in this forum's postings implied by the following comments .... but I am worried, perplexed and, frankly, aghast at the number of times I hear the phrases:
And, worst of all in my book:
It's this sort of apathy, tolerance, understanding and blind acceptance by potential and existing customers that certain manufacturers take advantage of (by consistently shipping products which simply do not do what it says on the tin) and which is now exploited as the norm!
And please don't get me wrong, it's not only CME ... it's just the UF8 is my latest bug-bear!
Such charlatans are probably in breach of all types of "sale of goods" and "fit for purpose" contractual law (sorry, guys, I'm British, so these phrases are actual vagaries of the English Legal System ... but I'm sure you know what I mean) which is why, in the UK, the vast majority of retailers, dealers and distributors will actually take these products back with little complaint and willingly offer refunds and/or replacements.
However, my principle concern here is that we're all helping to propagate this state of affairs: where we have become de facto (and unpaid!) beta-testers of such products and where the myriad forums replace the manufacturers'/developers' "Quality Control" work-benches. If this is the cost model manufacturers wish to adopt, then instead of charging us a premium for being early adopters of their products - providing them with the necessary start-up capital or free cash-flow to help them continue to develop and/or manufacture such products such that they actually work - then maybe these guys should allow us to acquire their products at a reduced price until such time as they actually work and then charge the full RRP! Perhaps that will help to ensure problems are fixed in a timely manner?
Yes, I understand this is all one of degrees and that the logistics of this particular thought experiment are manifest and legion, but I cannot get my head around the fact that as full-paying customers (or simply paying customers) we have to accept stuff that simply doesn't work and are fast-becoming complicit in ensuring version 1 of "Widget XYZ" doesn't work! Or rather, we don't have to ... but we choose to!
I'm sorry, but I have a day job and all I want to do with my kit is make music. If I wanted to be a software developer, industrial designer, or product technician, that's what I'd have trained for and that's what I'd be paid for! I'm a musician who just happens to require lots of electrons bumping into each other in fancy ways in order to create my masterpiece!
If you're one of the lucky few to earn plenty of cash (or derive the majority of your income) from writing / playing music and/or are one of those artistes who has to be at the cutting edge, then perhaps it's understandable that the manufacturers want your feedback and involvement in developing their latest gadget; however, the vast majority of us are not. And, if truth be told, we rarely need (or use) much more than the perfunctory feature set our gear provides, once that initial buzz and exploratory fascination has subsided.
If Korg, Roland or Yamaha - or even CME - want to offer me a free bit of tin on the understanding that I'll aid them in its development in a professional capacity, then fine. I'll willingly become one of their advocates and brand ambassadors down my local 100-seater pub.
However, I object most strongly to performing this invaluable role gratis ... without so much as a "thank you", and resent being taken for granted. And so should you!
There are some things manufacturers/developers readily acknowledge are entirely unacceptable to their consumers and woe betide them caught trying to take advantage of any latent good will; but musicians can be taken for a ride - and all-too-frequently we are - because we're a soft touch. We actively participate in allowing all this to happen by appearing grateful that at least 80% of what's promised does function.
Isn't it a weird dichotomy of modern living that mass market consumer products work and work well - almost without question - for long periods of time, often under extreme circumstances (have you ever actually looked at a builder's kettle!?) whilst costing very little; whereas our expensive (pseudo-niche) products are expected to be bug-ridden and fail all-too-frequently during normal use with ne'er a whiff of protest from their owners?
Steinberg actually announced an upgrade during the Cubase SX3 product life-cycle to deliver on promised features and to fix a plethora of long-outstanding bugs. Then they reneged on this and advocated upgrading to Cubase SX4 instead - a cost option, surprise, surprise - with many useful features once present in SX3 removed (not the first time they've actually reduced functionality during an 'upgrade'), many more bugs present than issues fixed, a wholly new (confusing and enforced) way of achieving certain tasks that many dislike, and a lack of assured backwards compatibility.
But then Steinberg have always done this and I'm one of those for whom the migration to another platform would be fraught with problems, something Steinberg no doubt rely on (see, I'm as daft as the rest of you!). To add insult to injury, one of the long-outstanding features which users have been requesting for years, is still not present (side-chaining)! There were murmurs on the Cubase Forum ... but now even that's subsided.
I'm all for adopting a pragmatic viewpoint if the problem is minor, specific or a simple oversight - being more tolerant if the solution really is 'just around the corner' and, importantly, a no cost option - but here we have a keyboard which:
And I certainly don't want to be told to change my working practices to fit in with what the keyboard is and is not capable of doing - I want USB to work reliably without reverting back to MIDI: that's why I bought it!
But what annoys me more than anything, is being told that the hardware is being 'upgraded' as a cost option to the customer ... when what is actually happening is the existing solution does not work (has never worked and often cannot be made to work) ergo the upgrade approach is being invested in by the manufacturer as ITS lowest cost route to resolving as many of the outstanding issues as possible. In fact, I'd guess that many of these so-called 'low cost' solutions are actually profitable for the manufacturer, which is why outstanding issues are resolved with an 'upgrade'!
So, let's recap: the CME forums are getting excited by an 'upgrade' ... which is actually being used to provide facilities WE'VE ALREADY PAID FOR.
Am I really alone in finding this kind of sharp practice morally reprehensible - not to mention quite probably illegal in many countries, if anyone had the time, money and wherewithal to push a test case through good ole jurisprudence - or is it that everyone has resigned themselves to the fact that 'this is just the way it is; don't fight it'?
The UF-series could've been great master keyboards, something to launch CME amongst the mainstream to compete with products from the big boys: Korg, Roland and Yamaha; as it stands - and despite its attractive price - it has its (tolerant) users, its red end-cheeks and a group of onlookers wondering why on earth we're sitting here bemoaning that the darned thing still doesn't work, laughing at us as we pay once again to achieve what should have been achieved several years ago at its launch!
Next time something doesn't work, or doesn't do what it initially promised to, spend the 2 hours or so driving it back to the dealer and get something that does, even if that means sacrificing on the look and feel elements - do red end-cheeks really compensate for stuck or missing notes? - or losing those super-duper 'new-improved' features you'll never really use anyway! This is the only way some manufacturers (and their distributors) will learn not to ship shoddy product and start testing it properly before taking our hard-earned cash!
[After all - and as an aside - what do all those in-country distributors actually do for their profit margins anyway? Very little in some cases, it seems, and certainly not much in the way of product testing, per se, as it appears they can leave specialist, exhaustive and costly testing to the magazines and early adopters. You know who you are ... and we're going to catch up with you! There are, of course, some superb examples out there ... but perhaps we should name and shame those 'distributors' which offer little more than a warehouse before marking up all that kit we crave!]
I'll even wager that taking the time to return under-performing kit will be far cheaper and much less time-consuming than investing in all the countless pizzas, Cokes, coffee, fags, takeaways, late nights, sore 'forum eyes' and fraught relationships caused by sitting in front of manufacturer forums helping them to find that workaround until such time as they can develop the requisite 'upgrade' required to fix the problem.
Cynical? Moi?
Rant over! Lunch-break finished.
Comments welcomed ... encouraged, even!
[OK, I've actually got a 'free' lunch-break, so time to rant - Ed!]
There's absolutely no criticism of anybody involved in this forum's postings implied by the following comments .... but I am worried, perplexed and, frankly, aghast at the number of times I hear the phrases:
- "being realistic";
"adopting a pragmatic standpoint";
"managing";
"understand its shortcomings";
"working around its shortcomings";
"work around the limitations";
"nothing is perfect in this world";
And, worst of all in my book:- "being realistic".
It's this sort of apathy, tolerance, understanding and blind acceptance by potential and existing customers that certain manufacturers take advantage of (by consistently shipping products which simply do not do what it says on the tin) and which is now exploited as the norm!
And please don't get me wrong, it's not only CME ... it's just the UF8 is my latest bug-bear!
Such charlatans are probably in breach of all types of "sale of goods" and "fit for purpose" contractual law (sorry, guys, I'm British, so these phrases are actual vagaries of the English Legal System ... but I'm sure you know what I mean) which is why, in the UK, the vast majority of retailers, dealers and distributors will actually take these products back with little complaint and willingly offer refunds and/or replacements.
However, my principle concern here is that we're all helping to propagate this state of affairs: where we have become de facto (and unpaid!) beta-testers of such products and where the myriad forums replace the manufacturers'/developers' "Quality Control" work-benches. If this is the cost model manufacturers wish to adopt, then instead of charging us a premium for being early adopters of their products - providing them with the necessary start-up capital or free cash-flow to help them continue to develop and/or manufacture such products such that they actually work - then maybe these guys should allow us to acquire their products at a reduced price until such time as they actually work and then charge the full RRP! Perhaps that will help to ensure problems are fixed in a timely manner?
Yes, I understand this is all one of degrees and that the logistics of this particular thought experiment are manifest and legion, but I cannot get my head around the fact that as full-paying customers (or simply paying customers) we have to accept stuff that simply doesn't work and are fast-becoming complicit in ensuring version 1 of "Widget XYZ" doesn't work! Or rather, we don't have to ... but we choose to!
- Sorry Mr BMW driver, we'll have the brakes and off-side lights working next month, just don't drive it too fast or at night.
Sorry Mr landlord, we'll have the Ethernet structured wiring and lifts operational in 6 months, don't use your computers and take the stairs.
Sorry about the crash, Mr train driver, we're only at release 1.2 of the traffic lights control software, it was to be expected. Could you go back to using a leather satchel for a while?
Sorry our microwave burnt your food, the firmware in the new-improved timer is still only at beta 0.9. Why not use your gas oven?
Sorry our MS Word software has lost your latest novel Ms Rowling but it is a Microsoft product after all! Perhaps you should use a typewriter.
I'm sorry, but I have a day job and all I want to do with my kit is make music. If I wanted to be a software developer, industrial designer, or product technician, that's what I'd have trained for and that's what I'd be paid for! I'm a musician who just happens to require lots of electrons bumping into each other in fancy ways in order to create my masterpiece!
If you're one of the lucky few to earn plenty of cash (or derive the majority of your income) from writing / playing music and/or are one of those artistes who has to be at the cutting edge, then perhaps it's understandable that the manufacturers want your feedback and involvement in developing their latest gadget; however, the vast majority of us are not. And, if truth be told, we rarely need (or use) much more than the perfunctory feature set our gear provides, once that initial buzz and exploratory fascination has subsided.
If Korg, Roland or Yamaha - or even CME - want to offer me a free bit of tin on the understanding that I'll aid them in its development in a professional capacity, then fine. I'll willingly become one of their advocates and brand ambassadors down my local 100-seater pub.
However, I object most strongly to performing this invaluable role gratis ... without so much as a "thank you", and resent being taken for granted. And so should you!
There are some things manufacturers/developers readily acknowledge are entirely unacceptable to their consumers and woe betide them caught trying to take advantage of any latent good will; but musicians can be taken for a ride - and all-too-frequently we are - because we're a soft touch. We actively participate in allowing all this to happen by appearing grateful that at least 80% of what's promised does function.
Isn't it a weird dichotomy of modern living that mass market consumer products work and work well - almost without question - for long periods of time, often under extreme circumstances (have you ever actually looked at a builder's kettle!?) whilst costing very little; whereas our expensive (pseudo-niche) products are expected to be bug-ridden and fail all-too-frequently during normal use with ne'er a whiff of protest from their owners?
Steinberg actually announced an upgrade during the Cubase SX3 product life-cycle to deliver on promised features and to fix a plethora of long-outstanding bugs. Then they reneged on this and advocated upgrading to Cubase SX4 instead - a cost option, surprise, surprise - with many useful features once present in SX3 removed (not the first time they've actually reduced functionality during an 'upgrade'), many more bugs present than issues fixed, a wholly new (confusing and enforced) way of achieving certain tasks that many dislike, and a lack of assured backwards compatibility.
But then Steinberg have always done this and I'm one of those for whom the migration to another platform would be fraught with problems, something Steinberg no doubt rely on (see, I'm as daft as the rest of you!). To add insult to injury, one of the long-outstanding features which users have been requesting for years, is still not present (side-chaining)! There were murmurs on the Cubase Forum ... but now even that's subsided.
I'm all for adopting a pragmatic viewpoint if the problem is minor, specific or a simple oversight - being more tolerant if the solution really is 'just around the corner' and, importantly, a no cost option - but here we have a keyboard which:
- has been plagued with all sorts of manufacturing faults from day one;
has repeatable problems defined on forums the world over;
has remained problematic for far too long with no end in sight;
is now going to cost us to put right.
And I certainly don't want to be told to change my working practices to fit in with what the keyboard is and is not capable of doing - I want USB to work reliably without reverting back to MIDI: that's why I bought it!
But what annoys me more than anything, is being told that the hardware is being 'upgraded' as a cost option to the customer ... when what is actually happening is the existing solution does not work (has never worked and often cannot be made to work) ergo the upgrade approach is being invested in by the manufacturer as ITS lowest cost route to resolving as many of the outstanding issues as possible. In fact, I'd guess that many of these so-called 'low cost' solutions are actually profitable for the manufacturer, which is why outstanding issues are resolved with an 'upgrade'!
So, let's recap: the CME forums are getting excited by an 'upgrade' ... which is actually being used to provide facilities WE'VE ALREADY PAID FOR.
Am I really alone in finding this kind of sharp practice morally reprehensible - not to mention quite probably illegal in many countries, if anyone had the time, money and wherewithal to push a test case through good ole jurisprudence - or is it that everyone has resigned themselves to the fact that 'this is just the way it is; don't fight it'?
The UF-series could've been great master keyboards, something to launch CME amongst the mainstream to compete with products from the big boys: Korg, Roland and Yamaha; as it stands - and despite its attractive price - it has its (tolerant) users, its red end-cheeks and a group of onlookers wondering why on earth we're sitting here bemoaning that the darned thing still doesn't work, laughing at us as we pay once again to achieve what should have been achieved several years ago at its launch!
Next time something doesn't work, or doesn't do what it initially promised to, spend the 2 hours or so driving it back to the dealer and get something that does, even if that means sacrificing on the look and feel elements - do red end-cheeks really compensate for stuck or missing notes? - or losing those super-duper 'new-improved' features you'll never really use anyway! This is the only way some manufacturers (and their distributors) will learn not to ship shoddy product and start testing it properly before taking our hard-earned cash!
[After all - and as an aside - what do all those in-country distributors actually do for their profit margins anyway? Very little in some cases, it seems, and certainly not much in the way of product testing, per se, as it appears they can leave specialist, exhaustive and costly testing to the magazines and early adopters. You know who you are ... and we're going to catch up with you! There are, of course, some superb examples out there ... but perhaps we should name and shame those 'distributors' which offer little more than a warehouse before marking up all that kit we crave!]
I'll even wager that taking the time to return under-performing kit will be far cheaper and much less time-consuming than investing in all the countless pizzas, Cokes, coffee, fags, takeaways, late nights, sore 'forum eyes' and fraught relationships caused by sitting in front of manufacturer forums helping them to find that workaround until such time as they can develop the requisite 'upgrade' required to fix the problem.
Cynical? Moi?
Rant over! Lunch-break finished.
Comments welcomed ... encouraged, even!
Stephen E Crane
-
- KVRer
- 25 posts since 15 Sep, 2005
I should add that I just spent some more time today leaving the UF8 and the DAW up during very long breaks, and nothing blew up. I did change some services and settings earlier. So it's seemingly some specific combination of settings, combined with the idle state, that's toxic to the CME driver, but still a bit of a mystery. Hmm, more tests. I have never lost notes on trills, and here again it could be I've optimized and stripped out just the right competing processes. I have required a MIDI all-notes-off once in a while (but what's new in MIDI-Land?), and I do reset the ASIO and MIDI within the DAW every now and then, takes seconds and keeps latency issues down. So, UltraJv, I'm crossing my fingers you keep trying stuff, and get mo' satisfaction.
I should add for anyone still asking the "which one is best" question that none of these MIDI controllers fully emulate a real piano action, it's REALLY down to personal preference. Some users will benefit most from a simple and inexpensive above-average spring-loaded synth action, like an Edirol, others without a need for aftertouch and wheels, can use the relatively low end "stage pianos". Only the M-Audio Pro88 was so sloppy I found it unusable. The UF8 is a great deal, if you need 88 keys, you want aftertouch and CC inputs, a few assignable pots and faders, you aren't arbitrarily opposed to running it through a MIDI interface if you did develop USB issues, and you find it feels good when you play, like your heart hooks up to your fingers and you forget what you set out to do.
Let's see how the new CME products fare, people let us know!
I should add for anyone still asking the "which one is best" question that none of these MIDI controllers fully emulate a real piano action, it's REALLY down to personal preference. Some users will benefit most from a simple and inexpensive above-average spring-loaded synth action, like an Edirol, others without a need for aftertouch and wheels, can use the relatively low end "stage pianos". Only the M-Audio Pro88 was so sloppy I found it unusable. The UF8 is a great deal, if you need 88 keys, you want aftertouch and CC inputs, a few assignable pots and faders, you aren't arbitrarily opposed to running it through a MIDI interface if you did develop USB issues, and you find it feels good when you play, like your heart hooks up to your fingers and you forget what you set out to do.
Let's see how the new CME products fare, people let us know!
-
- KVRer
- Topic Starter
- 21 posts since 3 Jul, 2004 from Staffordshire
peakdesign,
Would you mind if I took advantage of your willingness to resolve this matter for (and by) yourself by asking what bits'n'pieces you've actually "stripped out" thus far?
I'm one of those "arbitrarily opposed" to MIDI!
I don't want any damned Eighties technology solving my problems for me.
I'll continue to suffer at the hands of this new-fangled USB stuff.
Cheers ...
PS Am I one of the few laggards out there who has only just discovered the wonders of "MIDI-OX"? I'm just glad it's not called "USB-OX" as it would probably not work as well as this amazing utility!
All thanks go to Martin Walker of "Sound on Sound" magazine http://www.SoundOnSound.com for dragging me back to basics.
Would you mind if I took advantage of your willingness to resolve this matter for (and by) yourself by asking what bits'n'pieces you've actually "stripped out" thus far?
I'm one of those "arbitrarily opposed" to MIDI!
I don't want any damned Eighties technology solving my problems for me.
I'll continue to suffer at the hands of this new-fangled USB stuff.
Cheers ...
PS Am I one of the few laggards out there who has only just discovered the wonders of "MIDI-OX"? I'm just glad it's not called "USB-OX" as it would probably not work as well as this amazing utility!
All thanks go to Martin Walker of "Sound on Sound" magazine http://www.SoundOnSound.com for dragging me back to basics.
Stephen E Crane
-
- KVRer
- Topic Starter
- 21 posts since 3 Jul, 2004 from Staffordshire
Nah, not really - I'm the boss!
And anyways, I type real quick: it's a by-product of contributing so frequently to all these musicians' forums, trying to develop products for wayward manufacturers!
You ought to see me when I'm really frustrated!
And anyways, I type real quick: it's a by-product of contributing so frequently to all these musicians' forums, trying to develop products for wayward manufacturers!
You ought to see me when I'm really frustrated!
Stephen E Crane
-
- KVRer
- 25 posts since 15 Sep, 2005
Steve, none of this computer stuff works 100%, it's regrettable, but as an engineer I understand why, and we don't toss it all because there are no better alternatives for our present situation. Look at Cubase in your case, the learning curve for changing and the still-usable software keeps you a Steinberg user. That's all I meant by "being realistic". Besides, there's nothing immoral about understanding the extraordinary difficulty of making perfect drivers.
You should by all means lambast CME, but since in my case the pro/con ratio is not as doubtful, I'm more interested in figuring out how to narrow down the causes of these inconsistent problems. Of course if a class-compliant ROM/board became available, I'd "demand it" cheap or free, so keep hollering.
You should by all means lambast CME, but since in my case the pro/con ratio is not as doubtful, I'm more interested in figuring out how to narrow down the causes of these inconsistent problems. Of course if a class-compliant ROM/board became available, I'd "demand it" cheap or free, so keep hollering.
-
- KVRer
- 25 posts since 15 Sep, 2005
Steve, I started out with a Media Edition 2005 HP Dual AMD. I have been killing services very gingerly and looking at the event logs. I just killed all terminal services support today, and related stuff. The machine should not be going into "suspend" or triggering off any network activity. I can't tell you exactly what is doing what. There is NO missing notes problem, and maybe the Steiny crash is now reduced too, we'll see, but then again I had lost Steiny overnight before I got the UF8, so if anything it's just been more frequent with the UF8 driver. I'm still stripping services, I'll let you know when I feel it's bottomed out. Check if you don't have a connection problem on the USB, try another cable and use some alcohol to wet the contacts, then fornicate the connection several times.
-
- KVRAF
- 6478 posts since 16 Dec, 2002
brilliant news. my UF7 shares all the USB problems with the rest of you. luckily the MIDI port itself is 100%.bobsled wrote:http://www.cme-pro.com/forum/viewthread ... a=page%3D1Quite tasty indeed. If I don't choke on the price, whatever that may be. btw I haven't been holding my breath on this one. Due maybe this fall?#1. This is a hardware update. Not a ROM replacement. The mainboard of the old UF gets replaced with a UFv2 mainboard.
#2. The UFv2 mainboard is a USB class compliant device and will require no driver install on Macs and PCs.
#3. The UFv2 has a flash ROM, it will be updatable with new firmware and settings over USB.
#4. The UFv2's firmware is quite different, the function of many of the buttons on the face of the keyboards will change. This will require an overlay or just use a Brother P-Touch.
#5. In normal mode, transport commands send CC messages (CC115, CC116, etc.) over both USB and MIDI. MIDI Beat Clock and MMC commands have been removed.
#6. Some U-Ctrl features from the VX line are being implemented. Not all of them, there aren't enough buttons and faders, but several important functions.
#9. There are 9 default Velocity Curves where the old UF had 4.
#8. There are about 10 memory locations for saving your UF setups called Registration Banks.
-
- KVRer
- 25 posts since 15 Sep, 2005
Many USB midi drivers I've tried will not recover from a temporary disconnect from the DAW. I.e. the device disappears. So if the CME boards have bad/cheap connectors, it would produce all sorts of problems. Please inspect and clean, try another cable, etc. CME needs to do something about VISTA, so they have to change firmware on all their gear anyway, even if for no other reason. So let's make sure the oddball problems some of us are having more of than others are not simply due to an intermittent connection.
-
- KVRer
- Topic Starter
- 21 posts since 3 Jul, 2004 from Staffordshire
Agreed!
Another advantage USB has over MIDI - its tolerance of cheap cables and dodgy connectors!
Yet another huge leap forward for mankind.
Now, where did I put my can of Caig "DeOxit"?
Isn't it funny how musicians tend to find chemicals are a good source of stress-relief when all else fails?
Bl00dy expensive; bl00dy good ... and the can doesn't have a USB port in sight!
PS Just one clarification: when you said that you found the combination of DAW's and USB MIDI drivers to be intolerant of intermittent disconnections, have you ever found ANY DAW/driver combo which tolerates even a brief disruption? I'm genuinely interested to know as none of my other USB hardware (class compliant drivers or not) is recognised by Cubase, V-Stack, SAVIhost or any of the standalone VSTi's I've got. Only MIDI-OX seems able to cope! What are your experiences? I'm sure there are legitimate reasons for this issue ... but there must be developers clever enough to find real-world workarounds?
Another advantage USB has over MIDI - its tolerance of cheap cables and dodgy connectors!
Yet another huge leap forward for mankind.
Now, where did I put my can of Caig "DeOxit"?
Isn't it funny how musicians tend to find chemicals are a good source of stress-relief when all else fails?
Bl00dy expensive; bl00dy good ... and the can doesn't have a USB port in sight!
PS Just one clarification: when you said that you found the combination of DAW's and USB MIDI drivers to be intolerant of intermittent disconnections, have you ever found ANY DAW/driver combo which tolerates even a brief disruption? I'm genuinely interested to know as none of my other USB hardware (class compliant drivers or not) is recognised by Cubase, V-Stack, SAVIhost or any of the standalone VSTi's I've got. Only MIDI-OX seems able to cope! What are your experiences? I'm sure there are legitimate reasons for this issue ... but there must be developers clever enough to find real-world workarounds?
Stephen E Crane
-
- KVRer
- 25 posts since 15 Sep, 2005
Whoa, steve, are you saying NONE of your USB hardware is working well and you thought the UF8 was funky? Or only that none of it will tolerate a glitch?
Have you tried BTW all the other USB ports on the motherboard? I've seen many motherboards in which the video slot and the first PCI slot AND some of the USB use the same IRQ and interfere. My sense is that the CME driver is relatively primitive. It's easier to write such a driver than to write one that recovers easily from disconnection, from CPU suspension, from interference by other hardware. Besides, the DAW software is not on the whole fault-tolerant either. There's code one can write that works OK if all is well, and there's code that's designed to withstand problems and partial failures. The code that runs an automobile, a medical device or a military system is in the latter category. That's called fault-tolerance. It's not easy to design such robust systems. Look at what happened to the Russian triple-redundant computer system on the space station! Sure there were three CPUs, but they all were fed from the same power bus with identical delays, and when power glitched, because of the load when all 3 were switched on at once, they all 3 defensively shut off. Oops. Needless to say when DAW code that is incompletely debugged to begin with, instead puts a priority on deliberately failing, in response to suspicion of copy protection violation for instance, it's pulling in the opposite direction! The idea of further tacking on challenges to USB dongles strikes me as ridiculous, given that nothing was stable to begin with. USB might have a fault-tolerant communications protocol, but if neither the driver nor the application make use of recovery, it does not help. When you can unplug and plug back in a DVD burner without ruining the current burn, that's impressive, and some equipment is like that. But I've been surprised at how fragile both the MIDI USB drivers and the DAW applications are.
The Windows device driver model is evolving, and it takes 100 times more skill to write a fault-tolerant driver. It's not been a requirement, very few board drivers are. But I thought USB was designed for hot-swap! Yet I unplug any USB keyboard and it disappears from the MIDI list, and the whole system gets confused, often requiring a reboot. Edirol seems to have a more robust driver than most. But with any temporary hardware problem most DAWs have no way to recover. Seems that if a VST loses connection to a network drive the whole system ca be compromised, as drivers then sit in a forever wait, even if you force-kill and restart the hung app. It's pathetic. That's why I am not convinced that the CME driver is particularly at fault. That's what I meant by being realistic.
And so I think that people who are having USB problems need to check all the physical and software links that tie this tottering dungpile together, check, clean/cramolin the connectors, turn off as many background processes as practical, try different USB ports, see how many devices are on a given IRQ, and look for improvements, not a lightning-flash transformation from a house of cards into a bunker.
Maybe if one minimizes the surrounding challenges, the 1.05 CME driver works OK, and it is not "faulty", just not any more fault-tolerant than the rest of the software, maybe a little less so.
In response to your posts I'm bona fide trying to make my UF8 skip notes or do weird things with velocity, and it hasn't yet. But I don't want to develop a bad relationship with this board, and it's easy once you start looking for flaws.
For the future, using the standard "class compliant" model can help, if the DAW incorporates auto-recovery, but that's an if. Keep us informed, Steve, please.
Have you tried BTW all the other USB ports on the motherboard? I've seen many motherboards in which the video slot and the first PCI slot AND some of the USB use the same IRQ and interfere. My sense is that the CME driver is relatively primitive. It's easier to write such a driver than to write one that recovers easily from disconnection, from CPU suspension, from interference by other hardware. Besides, the DAW software is not on the whole fault-tolerant either. There's code one can write that works OK if all is well, and there's code that's designed to withstand problems and partial failures. The code that runs an automobile, a medical device or a military system is in the latter category. That's called fault-tolerance. It's not easy to design such robust systems. Look at what happened to the Russian triple-redundant computer system on the space station! Sure there were three CPUs, but they all were fed from the same power bus with identical delays, and when power glitched, because of the load when all 3 were switched on at once, they all 3 defensively shut off. Oops. Needless to say when DAW code that is incompletely debugged to begin with, instead puts a priority on deliberately failing, in response to suspicion of copy protection violation for instance, it's pulling in the opposite direction! The idea of further tacking on challenges to USB dongles strikes me as ridiculous, given that nothing was stable to begin with. USB might have a fault-tolerant communications protocol, but if neither the driver nor the application make use of recovery, it does not help. When you can unplug and plug back in a DVD burner without ruining the current burn, that's impressive, and some equipment is like that. But I've been surprised at how fragile both the MIDI USB drivers and the DAW applications are.
The Windows device driver model is evolving, and it takes 100 times more skill to write a fault-tolerant driver. It's not been a requirement, very few board drivers are. But I thought USB was designed for hot-swap! Yet I unplug any USB keyboard and it disappears from the MIDI list, and the whole system gets confused, often requiring a reboot. Edirol seems to have a more robust driver than most. But with any temporary hardware problem most DAWs have no way to recover. Seems that if a VST loses connection to a network drive the whole system ca be compromised, as drivers then sit in a forever wait, even if you force-kill and restart the hung app. It's pathetic. That's why I am not convinced that the CME driver is particularly at fault. That's what I meant by being realistic.
And so I think that people who are having USB problems need to check all the physical and software links that tie this tottering dungpile together, check, clean/cramolin the connectors, turn off as many background processes as practical, try different USB ports, see how many devices are on a given IRQ, and look for improvements, not a lightning-flash transformation from a house of cards into a bunker.
Maybe if one minimizes the surrounding challenges, the 1.05 CME driver works OK, and it is not "faulty", just not any more fault-tolerant than the rest of the software, maybe a little less so.
In response to your posts I'm bona fide trying to make my UF8 skip notes or do weird things with velocity, and it hasn't yet. But I don't want to develop a bad relationship with this board, and it's easy once you start looking for flaws.
For the future, using the standard "class compliant" model can help, if the DAW incorporates auto-recovery, but that's an if. Keep us informed, Steve, please.
-
- KVRer
- Topic Starter
- 21 posts since 3 Jul, 2004 from Staffordshire
peakdesign,
Good afternoon - lunchtime once again!
To answer your question: "No, the only USB device I consistently have problems with is the UF8" ... and I've tried the usual stuff - using hubs (powered and un-powered), using different motherboard ports, better quality cables, shorter cables, de-installing and re-installing the drivers before swapping over to those alternative ports, even using each of the USB ports (external AND internal) on a cheapo 5-port USB / 2-port Firewire Safecom PCI card I installed for some strange reason (DON'T buy them, they're simply not supported by their support staff!) - but it makes no difference.
Once in a blue moon the Syncrosoft key is not identified by my Korg Legacy set-up (and now-and-again by Cubase but, thankfully, never during a 'session') ... however, that's a problem which occurs so infrequently as to be considered 'normal'!!!
Whilst I welcome the introduction of class compliant drivers for the entire CME product range, I currently have one outstanding issue with such drivers - and perhaps this is simply down to the 4 specific devices I actually own, rather than being an inherent limitation of the driver developer or Microsoft's Windows XP O/S (I'm no device driver developer, just a simple musician!).
Once an external physical device is correctly installed via a USB port, the generic names they're given and the associations (or mapping) between the connected physical devices and these system-allocated (and generic) device names actually change (even though these devices remain permanently connected to the same USB ports) as per the "Windows MIDI" panel of the "MIDI" tree displayed in Cubase's "Device Setup" window!
For example, the "USB Audio Device (1)" which once referred to my PreSonus FaderPort in one Cubase session can randomly become "USB Audio Device [2]" next time, and the Behringer BCF2000 and BCR2000 control surfaces (for instance "USB Audio Device (3)" or "USB Audio Device [1]") will also 'swap' their names with these devices as and when they feel like it!
To be completely honest, the Behringer BCx2000 devices were fairly stable in this respect, as were the CME UF8 and Korg Legacy MS20, until the Presonus device was added ... and then mayhem!
To compensate for this annoyance, I initially began renaming the generic "USB Audio Device xxx" title allocated to the 'problematic' Fader Port (as I always have for any and all generic devices which appear in Cubase's "Device Setup" panel, in order to remove a source of potential confusion) each and every time I opened Cubase. Because I have other generic hardware devices installed which I regularly access from within my DAW, as well as other devices I periodically loan from my local store on a trial basis which must be installed un-installed, coupled to being anally retentive about ensuring such things are 'tidy', this just seemed an obvious thing to do!
However, it's actually entirely misleading and leaves me further back than where I started because the names are no longer generic but specific ... albeit wrong! It does not seem to matter in which order I turn the devices on, or whether I turn them on before booting up the DAW either!
As I also had to identify and rename the generic USB entries for the Behringer BCR2000 and BCF2000 control surfaces, this 'DAW maintenance' - performed every time I open Cubase - fast became annoying as the 'naming convention' never seemed to 'stick' to the hardware after a re-boot. Unfortunately, the problem is not constrained to the naming process ... the allocation of USB device to its actual function (as presented in the Cubase "Device Setup" panel) is also 'lost'.
I'd be interested to know if this is confined to Cubase or whether class compliant drivers have a tendency to 'swap' around on other DAW platforms, or even under Windows itself!
Finally, the missing notes problem (since the upgrade to v1.05 of the driver) is much less frequent and only occurs during softly-played passages, so it's a 'manageable issue'.
Don't'cha just love technology!?
Good afternoon - lunchtime once again!
To answer your question: "No, the only USB device I consistently have problems with is the UF8" ... and I've tried the usual stuff - using hubs (powered and un-powered), using different motherboard ports, better quality cables, shorter cables, de-installing and re-installing the drivers before swapping over to those alternative ports, even using each of the USB ports (external AND internal) on a cheapo 5-port USB / 2-port Firewire Safecom PCI card I installed for some strange reason (DON'T buy them, they're simply not supported by their support staff!) - but it makes no difference.
Once in a blue moon the Syncrosoft key is not identified by my Korg Legacy set-up (and now-and-again by Cubase but, thankfully, never during a 'session') ... however, that's a problem which occurs so infrequently as to be considered 'normal'!!!
Whilst I welcome the introduction of class compliant drivers for the entire CME product range, I currently have one outstanding issue with such drivers - and perhaps this is simply down to the 4 specific devices I actually own, rather than being an inherent limitation of the driver developer or Microsoft's Windows XP O/S (I'm no device driver developer, just a simple musician!).
Once an external physical device is correctly installed via a USB port, the generic names they're given and the associations (or mapping) between the connected physical devices and these system-allocated (and generic) device names actually change (even though these devices remain permanently connected to the same USB ports) as per the "Windows MIDI" panel of the "MIDI" tree displayed in Cubase's "Device Setup" window!
For example, the "USB Audio Device (1)" which once referred to my PreSonus FaderPort in one Cubase session can randomly become "USB Audio Device [2]" next time, and the Behringer BCF2000 and BCR2000 control surfaces (for instance "USB Audio Device (3)" or "USB Audio Device [1]") will also 'swap' their names with these devices as and when they feel like it!
To be completely honest, the Behringer BCx2000 devices were fairly stable in this respect, as were the CME UF8 and Korg Legacy MS20, until the Presonus device was added ... and then mayhem!
To compensate for this annoyance, I initially began renaming the generic "USB Audio Device xxx" title allocated to the 'problematic' Fader Port (as I always have for any and all generic devices which appear in Cubase's "Device Setup" panel, in order to remove a source of potential confusion) each and every time I opened Cubase. Because I have other generic hardware devices installed which I regularly access from within my DAW, as well as other devices I periodically loan from my local store on a trial basis which must be installed un-installed, coupled to being anally retentive about ensuring such things are 'tidy', this just seemed an obvious thing to do!
However, it's actually entirely misleading and leaves me further back than where I started because the names are no longer generic but specific ... albeit wrong! It does not seem to matter in which order I turn the devices on, or whether I turn them on before booting up the DAW either!
As I also had to identify and rename the generic USB entries for the Behringer BCR2000 and BCF2000 control surfaces, this 'DAW maintenance' - performed every time I open Cubase - fast became annoying as the 'naming convention' never seemed to 'stick' to the hardware after a re-boot. Unfortunately, the problem is not constrained to the naming process ... the allocation of USB device to its actual function (as presented in the Cubase "Device Setup" panel) is also 'lost'.
I'd be interested to know if this is confined to Cubase or whether class compliant drivers have a tendency to 'swap' around on other DAW platforms, or even under Windows itself!
Finally, the missing notes problem (since the upgrade to v1.05 of the driver) is much less frequent and only occurs during softly-played passages, so it's a 'manageable issue'.
Don't'cha just love technology!?
Stephen E Crane