Support Forum

As this is an old topic describing pre-historic problems on pre-historic versions of FS, you should note here what version of FS you are using. If its a pre-historic version, please update to a new version.

Furthermore I've used the K8062 a pretty long time, pretty intensive. Found several really bad problems with it. (Hardware as well as software, I did a complete driver rewrite myself but could'nt fix failures of hardware...). But never ever had a situation like people describe here. So that tells me it might be the pre-historic version of FS is the source of the problem.

Also check the K8062.exe. One comes with FreeStyler (as I can remember though) and one is installed in the Windows path. Maybe one of the 2 is bad.

I have an Velleman K8062/VM116 about half of a year and always encounter a problems with this interface from FS versions 3.01 through 3.3.2.
The issues are next: when I start FS with K8062 connected - program often crashes or hang-up during start (at about 85% of a starting progress bar). And to restart FS I have to shut down a process K8062.exe in Task Manager.
So, when I connect K8062 after FS started, DMX output LED blinking and everything works BUT I see some jitter in scanners positioning (and sometimes on another channels). Furthermore FS works with K8062 VERY unstable - it often hangs-up or crashes (especially when actively using keyb shortcuts, or editing and trying to save some edited Seq's in Offline mode during CUE's run).
PS: Why keyb shortcuts so slow at reaction?

[Notebook ASUS F3Ja (2Gb RAM), Windows XP SP3 Home, FS 3.3.2. Interface Velleman K8062 (factory assembled) ]

Mike Vaghur wrote: PS: Why keyb shortcuts so slow at reaction?

The keyboard shortcuts aren't slow. The VM116/K8062 is.
This and every other problem you described is the consequence of Vellemans bad design of the VM116/K8062 hardware and driver. Wanna know all of it? Search for posts of me and the VM116/K8062 in this forum and you'll learn everything about this crappy device (especially this post). You'll also learn there's only one solution: buy another (better) interface. The world becomes a better place then.

Mike Vaghur wrote:The issues are next: when I start FS with K8062 connected - program often crashes or hang-up during start (at about 85% of a starting progress bar). And to restart FS I have to shut down a process K8062.exe in Task Manager.
So, when I connect K8062 after FS started, DMX output LED blinking and everything works BUT I see some jitter in scanners positioning (and sometimes on another channels). Furthermore FS works with K8062 VERY unstable - it often hangs-up or crashes (especially when actively using keyb shortcuts, or editing and trying to save some edited Seq's in Offline mode during CUE's run).

remco_k wrote:This and every other problem you described is the consequence of Vellemans bad design of the VM116/K8062 hardware and driver.

@ remco_k: so what you are saying is that FS crashing using this interface is not a bug in FS itself but caused by the interface and not worth to investigate further?

remco_k wrote:

mike Vaghur wrote:Why keyb shortcuts so slow at reaction?

The keyboard shortcuts aren't slow. The VM116/K8062 is.

@ Mike: you can verify this by trying shortcuts with another Interface

So, does it possible to work with this interface without headache if obtain a new drivers (and does anyone has it?) or it looks like I pissed away about 150 bucks for Velleman?

LJ_krede.dk wrote: @ remco_k: so what you are saying is that FS crashing using this interface is not a bug in FS itself but caused by the interface and not worth to investigate further?

Yes, saying that. I did that 'further' investigation already. The K8062.exe crashes. Not FreeStylers fault, and can't be solved nicely by FreeStyler either.

LJ_krede.dk wrote:

remco_k wrote:

mike Vaghur wrote:Why keyb shortcuts so slow at reaction?

The keyboard shortcuts aren't slow. The VM116/K8062 is.

@ Mike: you can verify this by trying shortcuts with another Interface

He does'nt need to verify that with another interface. In the past I did plenty of tests showing the VM116/K8062 is the problem.
But if Mike wants, he can just have a look at the show output screen and see that the output values does respond fast and that the VM116/K8062's output lags a long time behind that.
Now I'm not saying that its impossible that FreeStyler is the source of the slow response problem. But since a VM116/K8062 is used, its obvious to me that that kit is the source of the problem.

Mike Vaghur wrote:So, does it possible to work with this interface without headache if obtain a new drivers (and does anyone has it?) or it looks like I pissed away about 150 bucks for Velleman?

The driver is the least of your problem. The interface being way to slow is the biggest of your problem. And regardless of the driver, the interface remains slow. Always.
There's one way of getting it 'faster' and thats shrinking the number of used DMX channels and keeping most channels at value 0. A new driver won't come out I think, because it would not really make a big difference.
So yes, in my opinion a 150 bucks are wasted. Thats why I got rid of the damn thing and bought an Enttec USB DMX Pro. (150 bucks seems a lot to me though, I can remember buying the K8062 kit (to build yourself) for about €70,-)
But if you are using just 8 channels the VM116/K8062 would run fast enough though. For small setups this interface is "usable", but not scaleable.

Ok guys! I've got it! (cause later I'd saw remco_k's message at Velleman support forum with complete explanation [here http://forum.velleman.eu/viewtopic.php? ... 018&p=7771 ]. So I suppose, too, that the only salvation is to update a firmware of a "box" (but it's possible If anyone could have an original microcode and full schematic (and I don't think Velleman will propose it).
But I can't wait for it! Well, now I'm planing to sale my VM116 for a some small architecture light project and buy (or make by myself) some of ENttec interfaces (I've used it most early before, but sold it out).
Thanks everyone for help!!!

Ok so there seems to be a lot of different problems with the VM116 but is there no one on the forum that can suggest what I shall do to solve my specific problem with only 5-8 channels with output?

In addition to what I wrot in my previou post I can mention that the VM116 worked as it should the first time I installed it on the laptop ( used it several times without turning of the comuter) but the next day when I restarted the laptop the problem was there.

I have removed all file related to VM116, lightplayer, dmxcontrol etc from the computer (including manual cleanup in regedit (Windows Registry)) and then reinstalled it again. Nothing helped! The problem remains.

I reinstalled it with exctly the same files as I use on my desktop where everything works as it should.

/Johan

If you SetChannels(512) and then StartDevice() then you get 7 or 8 channels only that produce output.
Reverse the order StartDevice() first and then SetChannels(512) and you get 512 channels

Setting the Channels count before starting the device does nothing so you are rewarded only with the default 8 channels. Win7(64bit). JohnA

Greetings, and welcome this lurker to his first post.

Thank-you, thank-you. Takes bow and gets on with the show....

Just a heads-up on the velleman 8062. I have a clean room firmware update that works at FULL SPEED USB. It is up to 80x faster than the factory firmware (LOW SPEED only.)

Without any alterations to existing supporting software my new firmware is 10x faster by virtue of being a HIGH SPEED HID device. (One packet, or output report, every 1ms Vs one every 10ms for LOW SPEED HID).

To gain the additional 8X throughput requires software that is aware of the 64-byte buffer (HIGH SPEED) Vs the maximum 8-byte buffer that LOW SPEED allows.

Currently I am in the testing stage and if there is anyone who is interested in helping out and getting the firmware update sent to them in good faith that they will provide feedback, please feel free to contact me. This is not vaporware nor some far off notion that I'm "gunna" do.

Currently I am doing extended testing using the "vixen" software as it allowed me to use my own DLL plug-in driver and bypass the existing velleman software. I have also done basic testing with freestyler and some other software and it works out of the box in "legacy mode." (10x speed increase.)

Later I hope that the full 64-byte mode (with extended commands and options) can be incorperated into freestyler.

While I have not been able to get vixen to send 512 slots at 44Hz (the absolute maximum), it does happily churn away with 64 channels at 100Hz update. I have not tested in between 64 channels and 512 at this time but I expect that the results will be good.

This firmware contains a bootloader so that even if it is NQR (Not Quite Right) first time around or I wish to add hardware faders, adjustable timings or whatever, it can be updated from the comfort of your own home.

I posted a rather long and more technical post on the diy christmas forum here:

http://doityourselfchristmas.com/forums ... -k8062-...

Please feel free to contact me if you see that this could be useful to you. Like I said I'm looking for testers so genuinely interested people here is a chance for both of us.

i would be interested in this.

Thanks for the offers of assistance that have been posted and PM'ed to me.

A quick update and how I hope this will unfold.

The update requires a new firmware chip. You remove the old one and simply plug in the new one. For the few beta testers I will send (at my cost) the firmware chip. These chips are "in the mail" to me and I expect them this week. (However the last time I ordered from this supplier they were dishonest about having parts ready to ship. I don't much like that sort of thing...)

For use with freestyler only the chip is required and at this time freestyler only supports the 10x legacy mode. The idea is to create a sufficient user base so as to make it worthwhile for Raphael to include the new interface in a future release of freestyler. (Basically and for the most add 0x60 in front of the existing Velleman commands and use a 64-byte HID report, very simple!)

For beta testers to be able to update the firmware requires a Windows based PC. There is both a C# .NET GUI and a command line app written in C++ to do this.

In the end this firmware chip does not even require the Velleman and has nothing really to do with Velleman. Anyone can build a USB-DMX interface using it, a crystal, 5 or so caps a USB connector an RS485 interface chip and a hand full of resistors and get a 512 channel fully buffered, firmware timed BREAK, MAB etc (based on RDM timing specs) for way less than an off the shelf "open" FTDI based VCP driven unit. Anyway that is stage two.

I will be in touch again this week as soon as I find out the status of the chip delivery. In the meantime I need addresses of where I will be shipping these out to so please PM these to me and on the understanding that I expect feedback and may call on you for some assistance as I do not have a load of DMX gear. Only what I roll for myself at this time. Hey it's way cheaper! Have soldering iron will burn fingers!

Thanks again!

JTR

Just a quick update. Working is going ahead well but interleaved with other duties. There is a partly written wiki on the open DMX site here.http://opendmx.net/index.php/Velleman_K8062_Upgrade

I will add more soon. Thanks to Simon Newton of OLA fame for his interest and assistance.

The update chips have arrived and I am doing to final port from my hardware to the exact Velleman hardware. There are little details to resolve like working the LEDs and the methods of selecting legacy 8-byte mode and Full speed 64-byte mode etc. The devil is always in the detail.

Sadly, my efforts to recompile the existing Velleman software from their source code have come to nought. I was able to get the HID component (package) compiled and loaded in Delphi but I cannot see to work out what I need for the FastTime component. A tremendous pity as all that was otherwise required was the alteration of 8-byte chunks to 64-byte chunks and then any and all DMX software that uses the existing Velleman DLLs would have become fully compatible without the need for my custom HID driver.

I may take another look or ask around on Delphi forums as this is the ideal solution. Failing that I have a cross platform HID component written in C++ that perhaps I can modify. Sadly I am an assembler guru all out to sea with this HLL stuff.

The good news is that the firmware has shown itself to work faultlessly at 512 non zero channels @ 40 updates per second. From a Velleman (mock-up!). Who would of thought.

Thanks JTR for a regular news and updates for Velleman K8062/VM116!

I've read a wiki and have a question: if I have an K8062/VM116 with PIC18F2550 chip on board already, could I reflash it with firmware update or Velleman has protected it from updates (re-burned EEPROM write access) so I need a new chip?

Thanks.
Mike

Hi Mike,

Yes it has come to my attention that later k8062s have the PIC18F2550 on board. Yes these can be reprogrammed via a PIC programmer. The Velleman code is probably protected but that does not stop anyone reprogramming the chip. It can be erased fully. What it will prevent is a roll back to the native velleman code.

That is one thing to keep in mind. There are also other issues here for me in particular to address as I had wrongly assumed that the chip would need replacing outright and I would be supplying the chips at low cost given that I can buy them in bulk for somewhat less than the going price I may have been able to subsidize the disability pension I live on and slowly build up a portfolio of USB and DMX devices and get of it completely.

Now I am torn as to what to do. Several parties connected to Velleman (but I stress not Velleman) somewhat "misappropriated" code I wrote previously and there is a very ugly history here. Now I bet they know exactly who I am. I remember them too...

The minute I start sending of hex files to people will be the minute that these large companies, will take my code, reprogram there stock and possibly up their selling price boasting that THEY created this update. Like I said, ugly history....

Anyway, here is some GREAT NEWS! It appears to me that I have succeeded in my rewrite of the native Velleman software! This means that any and all DMX software that relies on the Velleman DLLs and EXE file will work natively with my firmware and even be backward compatible with the existing Velleman firmware.

I have optimized the algorithm used and it can be up 350% of the speed of the current software / firmware pair under some circumstances with only my software rewrite. Even so that is not enough to get realistic performance out of the k8062. It is only a start.

It does however solve some big issues. It means that Freestlyer (and all other supporting windows software) will gain the FULL advantage of my firmware without having to be rewritten and it takes care of the VID/PID issue as I really cannot be using Velleman's without permission.

This development has delayed the release of the firmware a little but it has shortened the total time it will take to get the whole scope of this project fully realized.

I will do a little more testing to be sure that I have all my ducks in a row and then release the new EXE and DLL file then the firmware in some form or another.

Thanks for your interest and I hope interested parties stay tune as there is more to come in a timely fashion. Something existing k8062 owners have been waiting for...

Hi
I am very interested in the upgrade for the K8062
Please send more information (and a upgraded chip would be nice )
Thanks
Glen

Ok all,

Just a quick update. The project has taken a back seat as the chip I ordered ages ago were only shipped to me last week. They still have not arrived. Meanwhile I have been seconded by dangerousprototypes to work on integrating the USB stack I ended up co-authoring into there IR TOY.

Sorry for my lack of communication. I tend to drop everything and work on one thing at a time. Needless to say that this project will get my full attention again shortly. There is some information on the opendmx wiki here:

http://opendmx.net/index.php/Velleman_K8062_Upgrade

Thanks to those who have emailed/PM'ed me I will get back to you all shortly.

-Jim

JTR wrote: The good news is that the firmware has shown itself to work faultlessly at 512 non zero channels @ 40 updates per second. From a Velleman (mock-up!). Who would of thought.

Been a long time since I was here at the forum.
But...
If you'd read (lots) of my previous posts, you may have noticed I've also put a lot efford in the K8062 to get it to work faster. At the end, I completely rewrote the driver with no success, dumped the K8062 and bought me an Enttec Pro.

Now, years later, I browse this forum, bumped into this thread. Here's someone who also digged really deep into it. JTR is his name. And it seems he has a solution for this terrible problem.
He should be rewarded!

Good work! Hope you get it completely sorted. Not for me (as I don't have the K8062 anymore) but for all those other users.

Hi everyone!

I know I don't strictly belong here with this question, but you fellas seem to be quite experts in this device, so if it's not a problem i'll ask here.

Warning, totally noob questions and problem coming up!

At first i'd like to say that i'm new to DMX512 in general. I bought and assembled a vm116/k8062, it is built with the newer chip.(I would like to mention, that Maplin's website had no indication whatsoever that this is a self-assembly kit so you could imagine my surprise) Currently i'm at the stage of trying to test if it works, but i seem to be stuck. Either i did something wrong and no signal is getting out, or the wrong commands are going out. Right now i have a zero88 AlphaPack with a par64 hooked to channel one connected to the interface. Stand alone test mode has no effect, neither does the test software.

The computer recognises the device, the "power" LED is continuously on, while the "DMX" LED is flashing.

How should i test it/what should i do to indetify what i'm doing wrong? I also have a bunch of Pulse LEDPar56s, and some LEDj pixel bars available if you think that's easier to test, i assumed that the "dumbest" would be best.

Thank you in advance for your help!