|
Ok and 90% don’t know what the difference is. Just trying to throw you guys a curve bell!
I'm pretty sure I would not like to live in a world in which I would never be offended.
I am absolutely certain I don't want to live in a world in which you would never be offended.
Freedom doesn't mean the absence of things you don't like.
Dave
|
|
|
|
|
Hi,
Soldering is not something I've done many times however, I do have a question that I'm not sure
can be answered. I can't find motherboard schematics on the device.
issue : Roland TD drums is USB 2.0 out / need 3.0 out .. 2.0 too slow
Is it possible to take out the USB 2.0 and install 3.0. Yes this would be a hack.
This probably will require a firmware update( reverse engineer ) but not real clear on how USB 2.0
handles i/o.
Not real clear on buffering or hardware requirements ( guessing controller dev probably just only
allows buffering that 2.0 can deal with which is fine. It's the speed of 3.0 that I'm looking for
)
The TD-25 TD-30 or TD-20( more pics of the 20 on google images ) may give you some idea ..
Any Ideas ?
|
|
|
|
|
Almost certainly not. USB3 is orders of magnitude faster than USB2, and it's very unlikely that the chipset silicon designed for the slow data rate could even begin to keep up at the higher speed, even if we ignore the other differences. And that's without the appropriate drivers...
And the chipset for USB2 has no idea how USB3 negotiation is handled, so a USB3 device plugged in would still only work at USB2. It's not really practical to try and change the chips themselves, most modern stuff is BGA (teeny tiny balls under the chip where you can't reach them) or fine pitch SMT (where you need specialist tools to desolder / solder without pulling the PCB to bits - Weller do a Flow-tip for some of their irons which can cope with 0.5mm pin pitch for assembly, but disassembly needs heatguns and a lot of care). Even if you had the kit, without circuit diagrams it would be a nightmare. Bear in mind that complicated PCBS tend to be multilayer as well, so just following tracks needs a damn X-ray machine...
Unless you are a seriously good solderer I'd have to suggest you look at upgrading kit if something's too slow - home alterations to modern PCB's are unlikely to succeed.
Bad command or file name. Bad, bad command! Sit! Stay! Staaaay...
AntiTwitter: @DalekDave is now a follower!
|
|
|
|
|
Thanks for the detailed answer ! Oh well 
( Pls ignore the comment above where I wrote replace 2.0 w/ 3.0 (hardware) .. Obviously, error checking would fail at the receiving end every time w/o firmware update )
These days acoustic drummers are using electric kits to practice. We are triggering drum software on laptops etc for extremely high quality ( studio recorded .. Superior Drummer 3 .. better sound than module ) drum sounds. As usual latency ( usb 2.0 / MIDI ) is an issue along with win drivers ( asio4all driver helps ).
Anyway, just looking to see if my question was an option as 3.0 is much faster.
Thanks again !
|
|
|
|
|
I came across the same issue trying to run UVI Workstation from a MIDI connected keyboard. As you said, the asio4all driver helps but the latency is still noticeable...enough so that I lost interest and haven't played it in over a year. I still have a digital piano in the living room but was looking for something that could replicate my first keyboard, a Korg DW-8000.
I'm really not sure that the USB 2.0 connection is the cause of the latency, but rather suspect it might be due to issues with your sound card. Good luck! 
Edit: There was a time when I was able to plug a bass into a line-in on the sound card and jam with media player either through my nice computer speaker system or through headphones...with no noticable latency. When I moved to Windows 7, despite having better hardware, and despite using the asio4all driver, the latency issue made it impossible. The solution: Move my bass amp into my home office. One of these days, and soon, I intend to buy an external device dedicated for this purpose so I can go back to using the headphones.
"Go forth into the source" - Neal Morse
modified 20-Jan-18 14:47pm.
|
|
|
|
|
I agree with everything OG has said on this, but I would offer two further points to consider:
USB 2.0 can operate at 60 MBS (that's bytes not bits). Nyquist theory for A2D conversion -basically- says that sampling anything above 2 X Freq_max is good, suggesting that for audio, sampling above 40 KHz would serve the audio spectrum. You have lots of slack in 60 MBS.
The second point is that windows is time non-deterministic in events. That is, you do not know if USB events are 15 ms between, or 50 ms between, you have no capability of forcing this. And for the software to do the turn-around takes a bunch of time. To do time deterministic events, you would need a Real Time Operating System or RTOS like some versions of Linux or QNX or NI Labview RT.
That said, I don't think your problem is in the USB 2.0 area, but in some other area that is causing the delays.
Ken
|
|
|
|
|
I'm going with Griff on this one
While the USB port itself maybe a simple physical upgrade, all of the supporting infrastructure of wiring, controllers, and other chips going to need to be upgraded as well.
Once that is done custom coding is most likely going to be needed to get this to play nicely with the rest of the system.
And when that is all done you will most likely find out that the rest of the system doesn't have the bus speed and bandwidth.
Director of Transmogrification Services
Shinobi of Query Language
Master of Yoda Conditional
|
|
|
|
|
No. This spec extends from the USB connector, though the chipset, that open collector buffer thing all the back to the firmware implementation.
Are you certain the latencies you are experiencing are truly due the the speed of usb 2.0?
Your issues are likely your output sound setup (DAW etc...) and not usb 2.0. Chances are your notes
are getting to the output device on time and then being backed up by plugin delay compensation or ASIO buffer size etc. Various DAWs do tight midi to differing degrees. With word on the street that Cubase is best in this regard.
Remember: good old midi (5 pin DIN) was 110 baud. Were talkin iceberg slow. And every thing was good.
That speed can keep up with the fastest music any human may want to hear. USB 2.0 blows old midi away.
|
|
|
|
|
HarvestMoon0000 wrote: Roland TD drums is USB 2.0 out / need 3.0 out .. 2.0 too slow
What makes you think it is too slow?
|
|
|
|
|
|
Java continues to be very popular and in demand. Server devs tend to use IntelliJ and Android devs tend to use Android Studio. Eclipse is old hat - I don't know any devs who still use it.
/ravi
|
|
|
|
|
Ravi Bhavnani wrote: Eclipse is old hat - I don't know any devs who still use it.
Yes, I've been using Android Studio (IntelliJ) for my Android work also and that's what I thought.
And that is good to hear about Eclipse.
I'm downloading netbeans now to see what it is like. It is 221MB v the 2GB for JDeveloper.
Thanks
|
|
|
|
|
Ravi Bhavnani: Eclipse is old hat - I don't know any devs who still use it.
Yeah well I guess we don't know each other! 
We still use Eclipse for our older projects where I work. But just to make everything perfectly clear I'd never use Eclipse it's terrible.
|
|
|
|
|
IntelliJ IDEA is the gold standard. Large and active community as well as many relevant plugins.
|
|
|
|
|
Robert Nadler wrote: IntelliJ IDEA is the gold standard. Large and active community as well as many relevant plugins.
I figured that might be the case because it is very nice for Android work ala Android Studio.
I was looking for a _free_ option, because I am _cheap_. 
|
|
|
|
|
|
I had a customer for a while (a very large company known by a TLA) who used Java widely for their in-house manufacturing control systems. Their primary motivation was because it worked on all of the OSs they needed to support which included AIX, Linux, OS/2, WinNT, 2000, XP, and who knows what else. They were able to have a very, very high degree of reusability between all of those platforms also. If I recall correctly, that factor was 100% once the figured out the various issues they had. These were systems that communicated with PLCs, other control systems, and their manufacturing database. The primary function of their systems was to interface between the manufacturing database and the various manufacturing systems. I was involved in the development of several of those manufacturing systems. AFAIK, they are still in production too. That was a long time ago, in a galaxy far, far away.
|
|
|
|
|
That was a great story and a great example of the goodness of Java. I really like to hear stories like that about technology.
Thanks for sharing. 
|
|
|
|
|
Some of our projects are very similar: interconnecting PLC's, manufacture DB's, process control systems and getting the result to layer 4 or 5.
The choice of technology you have in such an environment is really limited, because you often need to support various OS technologies and each version is fairly old. For Windows systems in particular, we have to support Vista with no updates, except for security patches.
Instead of using Java, which would be the obvious choice, we package everything into .NET Standard 2.0 libraries. For modern systems we run those with .NET Core, for legacy systems (both Unix and Windows) we use a custom Mono branch. It's always weird and nice to see brand new nugget packages actually working on Vista.
|
|
|
|
|
Good question, shame there arent more answers.
So far it seems to be usable on Androids and that's it.
My own, outsiders opinion, it that it is dead. I had a friends some years ago who used Java, who agreed, except for J2EE, what ever that is.
.Net (a Javaesque VM), and C# (Javaesque language) pretty much killed it off after Msft were told to stop messing with the Java standard by taking so much market share.
|
|
|
|
|
Munchies_Matt wrote: So far it seems to be usable on Androids and that's it.
My own, outsiders opinion, it that it is dead.
I think I have to agree. It just seems as if it is irrelevant now as there are other newer better ways that are better supported to get what you want.
I only really got involved with Java because of Android and now Android may very well shift to Kotlin.
And these reasons we are mentioning are probably why Google is moving toward Kotlin for Android.
|
|
|
|
|
Not to mention avoiding another drawn out baseless lawsuit from Oracle Corporation.
|
|
|
|
|
I worked on a project years ago that was a mix of Java client and C/C++ on Linux server. We chose Java because the client needed to run on all platforms, every flavour of Unix/Linux, Windows, Mac etc. As one of the few Java guys I did most of my development in emacs, only switching to eclipse late in the project. I think I still preferred emacs.
|
|
|
|
|
Yeah Eclipse was so painfully bloated and slow. I was so glad when Android moved to IntelliJ Android Studio.
|
|
|
|
|
Well it was for Android, but not for ordinary Java development. I think the bloatishness is caused by the structure of Android applications, rather than the IDE. Early versions of Android studio were utterly useless on anything less than Performance Computer, XC Series Supercomputers - Cray[^].
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
