The Kronos Inner Workings - For the brave ones ;-)

Discussion relating to the Korg Kronos Workstation.

Moderators: Sharp, X-Trade, Pepperpotty, karmathanever

Locked
Bruce Lychee
Platinum Member
Posts: 856
Joined: Thu Jun 16, 2011 4:16 pm

Post by Bruce Lychee »

Sharp wrote:Hi Bruce.
Have a read of the anti circumvention laws.

Those laws in conjunction with KORG's Encryption and copyright statement protect KORG's work.

Regards
Sharp.

I'm well aware of IP laws in the US. Nevertheless, my point is that making general statements of law that may or may not be applicable to this particular case is not helpful on your part. Furthermore, moderators of this forum have shown in the past that applying sound legal analysis is less important than justifying the forum's policies... Which is fine. I just think it is misleading to be using the law as your justification unless you are in a position to properly analyze the laws and the facts of this case.
Shigeru Kawai SK5
Roland Jupiter 80
Vintage Vibe 64
roblof
Full Member
Posts: 102
Joined: Fri Dec 12, 2008 7:41 pm

Post by roblof »

danatkorg wrote:First, let's clear up the misconception that the Pa3X Oriental was developed by an independent third party.
Tnx for chiming in Dan! I'm happy to see you respond here.

This thread have taken a spin that I couldn't think of when I started it.

I was actually encouraged by that Pa3X-Oriental rumor :oops: and others efforts to find out the nature of our baby kronos and when the opportunity came up I truly believed in the community 'helping out' as have been done in other similar ways in other products.
This has resulted in a system that we're very proud of, with great-sounding algorithms and outstanding runtime efficiency
I am also proud of your efforts. This oasys/kronos is great work and that's why it is such great fun to add just that bit of extra stuff - The icing of the cake, so to speak :P .

From my point of view it's like having a BMW with a speed cap (unfortunetly they are capped that in my country) and I (with others) want to remove that cap. Now, I've already done that by the help of 3'rd party efforts being made. Does that analogy make any sense?
The idea that one or two people could successfully reverse-engineer and completely understand these binaries in a few weeks and then modify them successfully in a few months of part-time work, without the source code, and without unintended negative side-effects, is simply not credible.
The thought was not to have one or two people re-write the kronos os- and audio engines, but rather tweak and focus on specific stuff. This is much easier than doing major rework.
Because of this, it's not surprising that the third-party technical analyses posted in this thread have been rather incorrect.
That's the core nature of doing analysis. One tend to work from an theoretical and initial understanding and having to revisit the premature conclusions being made - That's the beauty of analysis. Sometimes you are right, and sometimes you are wrong along the way.

I'm now official 3'rd party :D
I'll go into one example. As we've said before, all synthesis, effects, and audio processing is done within the Intel CPU, and naturally the sequencer and KARMA as well. The third parties have proposed that this was not the case, and specifically suggested that changing the USB buffering scheme (currently carefully optimized for low latency - a fundamental musical concern) would somehow affect "rhythms" continuing across SST. One might note that rhythmic wave sequences, step sequences, and LFOs do in fact continue across SST, while KARMA and the sequencer/drum track stop, and easily conclude that this is unrelated to USB buffering.
I'm admiring you guys to pull this of. Especially by only using a "low-end" atom based resources.

This is also why it's so hard to believe that all this is being made on the mobo without the help of shark and therefor one start to speculate on "the inner workings".

As I somewhat said before, proposing isn't the same as doing. The initial effort of the usb was to see if there was a way to speed up the boot time in some way or another and based upon initial assumptions ideas always pops up.

You should only consider our published efforts to be more like a blog of progress along the way, rather than factual statements and I believe the community understands that.
Such high-level technical misconceptions by these third parties strongly suggest that they have overestimated their current readiness to begin modifications, and that it is unlikely that their promises of improvements would ever have been fulfilled.
This may, or may not, be true but now we'll never know...

About overestimation - Do I dare to mention the kronos editor?

I.e. I for one wanted the cx-3 to respond to two or even three midi channels instead of the twisted effort of range splitting the keyboard to have upper/lower manuals.

I also wanted to have the vga-output to sync lyrics/chords with my set-lists.

Such specific tasks should in theory be able to become fulfilled since they rely on functionality already there one way or another.
Many of the suggestions posted on this forum are on our wish lists as well. We are listening, and it is our goal to make instruments which are as useful to you as they possibly can be (and to us - we play them too!).
And we are like kids, waiting for what you may come up with the help of our wishes.
We're grateful to have you as customers and fellow KRONOS-users, as this wonderful instrument continues to evolve.
And I for one is proud of being part of the korg community since way back due to the great stuff you come up with.
Best regards,

Dan
Thank you Sir. All the best!
User avatar
SeedyLee
Platinum Member
Posts: 1397
Joined: Wed Sep 13, 2006 1:48 pm
Location: Perth, Australia

Post by SeedyLee »

Happy to have you join the fray Dan!

With respect to the use of the ATOM CPU vs ARM for sound generation:
danatkorg wrote: Because of this, it's not surprising that the third-party technical analyses posted in this thread have been rather incorrect. I'll go into one example. As we've said before, all synthesis, effects, and audio processing is done within the Intel CPU, and naturally the sequencer and KARMA as well. The third parties have proposed that this was not the case,
Heider wrote:
I am assuming that the EVA.img (and the other image) are some sort of EXT2/3 encrypted filesystem that is mounted for SHARK's use, I am getting closer to think that the Linux does not even decrypt these files nor decrypt-mount them for SHARK.
SeedyLee wrote: The ARM processor in the Kronos is too underpowered for the sound generation and effects duties of the KRONOS, and I specifically remember an article featuring an interview with Dan Philips himself where he mentions that one of the benefits of the dual-core Atom was using one core for EXis, and the other for effects.
.Jens wrote: Heider, with all respect for your investigations - but I am 99% sure that the atom CPU is not supporting a dedicated sound processor here, but is actually the core of the sound processing engine in fact. From several statements in forums like this and simple logic one can easily deduce that the ARM is doing most of the I/O stuff, while the atom(s) are doing the main part.
Also, the impact of various buffering techniques:
danatkorg wrote: and specifically suggested that changing the USB buffering scheme (currently carefully optimized for low latency - a fundamental musical concern) would somehow affect "rhythms" continuing across SST. One might note that rhythmic wave sequences, step sequences, and LFOs do in fact continue across SST, while KARMA and the sequencer/drum track stop, and easily conclude that this is unrelated to USB buffering.
Which mirrors my assessment back on page 10:
seedylee wrote: If the use of single-buffering was the cause of rhythms not being sustainable through program changes, why does the employment of a single buffer not affect the operation of SST?

Using a single buffer between SHARK and the motherboard surely doesn't preclude utilisation of more complex buffering techniques within software that would handle more complex functions such as SST - in fact, if this was the case, I would imagine that a single hardware buffer would be preferable from a latency and memory efficiency perspective?

My understanding is that using a double-buffering technique for USB transfer would mean that neither buffer could be touched whilst a buffer copy or buffer switch was occurring, potentially introducing latency. There could be a very good performance reasons that a single-buffer was employed.
I had my own assessment of the architecture based purely on published literature, logic and "best-practice" in system design, which was different to Heider's. I wasn't 100% confident in my analysis as I didn't reverse-engineer the system. This doesn't mean I though Heider was wrong, not did it diminish my appreciation of his highly technical efforts. But it did make me question how quickly such significant changes could be implemented at a binary level without substantially affecting the overal performance of the system. I was skeptical to be sure.

I also think it's a big call to suggest that the KRONOS is full of "bad code" when looking purely at binary objects, without any regard to many of the design decisions that went into its construction. Okay, some of the Linux scripts included on the restore CD were a little unpolished in the first release, but you know what? Korg have improved them in subsequent releases.

To suggest that Heider should be given a job at Korg for his work is noble, but consider that he hasn't actually (as far as we know) managed to implement any of these features, and many of his plans were based on incorrect technical analysis. Do I think he could implement these features if he had the source code? I have no doubt, but I also have no doubt that Korg could introduce these features too.

There's an infinite number of features that could be implemented, and only a finite amount of time, so it always comes down to selecting which features take priority. Not having features such as KARMA that sustains through program changes isn't a sign of incompetence or bad-coding practices, it's a sign that another feature was chosen ahead of it. It's Korg's product, they get to call the shots on which features are included and which are excluded. It would be nice (and probably commercially beneficial to Korg) if they listened to their customers, but they're under absolutely no obligation to do so. If you don't like it, buy another product! The problem with the KRONOS however is: which other product could one even consider buying as a viable alternative?

Once again, this doesn't diminish Heider's technical ability, but it does point out that it is really, really hard reverse engineering and modifying a complex, real-time system.
Current Equipment:
Korg Kronos 2 88, Reface CS, Roland JV-1080, TE OP1, Moog Subsequent 37, Korg ARP Odyssey, Allen & Heath Zed 18, Adam F5, MOTU MIDI Express XT, Lexicon MX200 & MPX1, Yamaha QY700, Yamaha AW16G, Tascam DP008ex, Zoom H6, Organelle, Roland J6 & JU06A

Previous: Triton LE 61/Sampling/64MB/4GB SCSI, MS2000BR, Kronos 1 61, Monotribe, NanoKontrol, NanoKeys, Kaossilator II, Casio HT3000, Roland VP-03, Reface DX, Novation Mininova, MPC One
Heider
Guest
Posts: 102
Joined: Thu May 17, 2012 1:07 pm
Location: Wimbledon
Contact:

Post by Heider »

Alright.... I've been out all day, and now I just read all this...

I've almost written a very large message, then changed it three times, then didn't like the tone on them,.. then I decided to wipe it all off and then write the following:


Ok, so, without attacking anyone, or sharing any negative comments,... let's get the record straight here...

1) Based on the emails I forwarded to Korg upon thier request; these emails shows that Korg (Mr. S.) mentioned there "no such thing called PA3X Oriental, and that a distributor developed their own USB stuff in Germany and that Korg knows nothing about it"... however Korg (R&D) recently mentioned "Actually, there is!,.. and oh yes, it was made by Korg modifying with different ROMs and options etc"... so, just wanted to remind Dan here, please note that it is unfair to even state that what I posted on the original thread regarding the PA3X was "a rumour", again, thanks to Mr. S here, I now have two story versions of the "PA3X oriental" keyboard is all about...


2) Korg is saying here that "We care about customers, and currently working on these changes already"... I'd say "Fantastic!, very nice, and saves my time"... Are you (Korg) willing to commit to a date whereas these changes are complete for this Kronos?


3) Are you (Korg) willing to give us a "list of changes with their ETAs" that you are working on now for your releases?


4) If you are un-willing to comment on date for fix, and you are un-willing to give any list of what you are working on (that is if you are working on these enhancement requests anyway), then are you willing to accept that "If I would to be given this task, then I certainly wouldn't mind?"...


5) Since it's very unfair from Dan to forbid me from working on a Kronos Update (because he fear's I might do it, what else?!)... and then states: "it's impossible that these parties would ever release anything in few months time, since they are talking complete non-sense about how the USB does it's buffeing based on the "Technical Early Speculation" discussion about how the USB does it's buffering without taking into account the fact that this thread started from scratch where (Heider) calling the ARM system (SHARK) and the Intel CPU/Mtherboard as (MOBO) and clearly started from scratch knowing nothing about it initally, but then in less than 6 days, managed to get to the holly code that people put 100 man-years of efforts to it)...

Fine;... Now, (Gloves Off!)... the question to you is (although it sounds to you that I'm out of my mind), if (just "what-if...") I (Heider, "alone") am willing to commit to all on date where a list of fixes are published, with an ETA to a final release, without disclosing any information to public whatsoever other than the final release itself,... Do you accept the challenge (Korg)?!...

And by the way,... if you accept, then I would challenge you to crack my encryption then,... I am not joaking here ("I SWEAR TO GOD"... I have done a much better encryption than your current one when I was in school,... that's school and not high-school by the way), when I was 16 years old, I wrote my first protected-mode operating system with a complete Intel 386-Insturcion-set-suport debugger, I'd like to believe that I was the first who created "my own" virtual machine under my OS just to debug things around and learn... (not share, but just learn how they work)... in all honesty,.. in some cases now, I don't need a disassembler to work as I can understand the code and change it by just looking at it's hex values... ! isn't that nice?


Stop trying to make me look like I don't know what I'm doing,... you know I do, and I have proved that I can,... in just under 6 hours, both EVA and MOD files were at the palm of my hand...

Common Korg, since you are posting here already, what do you say?!... fair challenge, I guarantee no harm would be made sine no code would be shared. At least your customers would love you for at becoming open about it... think about it, have you got anything to loose?!...

It would be a shame if you chicken out, or hide behind your license's T&C, or stop replying at this stage,... or even refuse to comment on your own ETA or list of fixes otherwise!

Consider this as my signed NDA if you like this to be a challenge... Heider "alone" vs. "Korg"... in 6 months time, who releases the better patch?!...



Yours,
Heider
Heider Sati
User avatar
apex
Platinum Member
Posts: 2340
Joined: Tue Feb 23, 2010 6:22 am
Contact:

Post by apex »

Heider...

this is my main question, what good could come of this? I mean if you are/were able to come up with a great patch.... what next? It's not like you could release it for all of us to use. and it's not like we would know how to do the things that you have done (to our own machines)

it's crazy to think that we could mail you our Kronos and have you go through this entire process with each of our machines.

what good is going to come of this other than you and us knowing what is possible and what you and ONLY you are able/capable/willing to do?

can any good come from this? It just seems like it could come to a mine is bigger than yours contest. (or you could get hired full time by Korg...) wouldn't that be something!!!
Please subscribe and follow me on Youtube and Facebook for information related audio technology.

YouTube -
https://www.youtube.com/juliusdeberryjr

Facebook -
https://www.facebook.com/majesticstudiosllc

Instagram - https://www.instagram.com/majesticstudios_jld/
User avatar
MarPabl
Platinum Member
Posts: 938
Joined: Sat Sep 24, 2011 12:09 am
Location: MX

Post by MarPabl »

+1 Heider :!: :!:

This thread has the sole purpose to create a better Kronos for everyone. There are plenty of examples of this like it has been stated before (just missing the jailbreaking on the iPod/iPad/iPhone or homebrew)

Korg is acting like Sony did and what did they got? Public shame showing indeed how poor their security is: PlayStation Network outage [-X
Last edited by MarPabl on Sun Jul 08, 2012 4:36 am, edited 1 time in total.
Current gear: :arrow: Access Virus TI2 Whiteout Keyboard (111/150), Access Virus TI2 Polar DarkStar Special Edition, Gibson Custom Lite 2013, Roland MV-8800 \:D/
User avatar
apex
Platinum Member
Posts: 2340
Joined: Tue Feb 23, 2010 6:22 am
Contact:

Post by apex »

what does +1 mean? lol
Please subscribe and follow me on Youtube and Facebook for information related audio technology.

YouTube -
https://www.youtube.com/juliusdeberryjr

Facebook -
https://www.facebook.com/majesticstudiosllc

Instagram - https://www.instagram.com/majesticstudios_jld/
User avatar
MarPabl
Platinum Member
Posts: 938
Joined: Sat Sep 24, 2011 12:09 am
Location: MX

Post by MarPabl »

Wow good point your won my privileged place just below the last Heider's post :?

We don't need to send the Kronos to Heider to apply the enhancements, he began by making space for a network jack, but it's possible to make the Kronos support WiFi just as we have on Motif XF.

I think that if we can get a better UI, that's great for everyone.
Current gear: :arrow: Access Virus TI2 Whiteout Keyboard (111/150), Access Virus TI2 Polar DarkStar Special Edition, Gibson Custom Lite 2013, Roland MV-8800 \:D/
User avatar
apex
Platinum Member
Posts: 2340
Joined: Tue Feb 23, 2010 6:22 am
Contact:

Post by apex »

oh ok. cool

but how could these enhancements be applied to all of our machines?

Korg is not gonna do it. and 9 our of 10 of us aren't gonna go it.
Please subscribe and follow me on Youtube and Facebook for information related audio technology.

YouTube -
https://www.youtube.com/juliusdeberryjr

Facebook -
https://www.facebook.com/majesticstudiosllc

Instagram - https://www.instagram.com/majesticstudios_jld/
User avatar
MarPabl
Platinum Member
Posts: 938
Joined: Sat Sep 24, 2011 12:09 am
Location: MX

Post by MarPabl »

It can be applied as a software upgrade, just as we upgraded to 1.6.0

Who knows about who's willing to apply the update... many members here have applied several fixes:
- Adding RAM memory.
- Changing the SSD.
- Changing the fan.
- Changing the sliders.
- Correcting sticky, clunky, noisy keys.
- Big etc. for the rest of fixes I may be missing.

I think many guys will give it a shot :) after all, here you can find how to restore your system to 1.0.4 by using the restore DVD. You have not too much to lose if you give it a shot :idea:
Current gear: :arrow: Access Virus TI2 Whiteout Keyboard (111/150), Access Virus TI2 Polar DarkStar Special Edition, Gibson Custom Lite 2013, Roland MV-8800 \:D/
User avatar
apex
Platinum Member
Posts: 2340
Joined: Tue Feb 23, 2010 6:22 am
Contact:

Post by apex »

you don't think it could brick your unit? and he's using the USB connection to attach network interfaces and connect the Kronos to Wifi huh?

seems interesting. very brave move I might add. Who's got $3000 to waste if you screw up your system... not me lol.

I do take my hat off to Heider for having this type of confidence in his abilities though.

it will be interesting (if nothing else) to see how all this turns out.
Please subscribe and follow me on Youtube and Facebook for information related audio technology.

YouTube -
https://www.youtube.com/juliusdeberryjr

Facebook -
https://www.facebook.com/majesticstudiosllc

Instagram - https://www.instagram.com/majesticstudios_jld/
User avatar
apex
Platinum Member
Posts: 2340
Joined: Tue Feb 23, 2010 6:22 am
Contact:

Post by apex »

I wonder what someone like him could have done to improve the Roland Fantom G. it was a system update away from being just about untouchable in a lot of aspects.
Please subscribe and follow me on Youtube and Facebook for information related audio technology.

YouTube -
https://www.youtube.com/juliusdeberryjr

Facebook -
https://www.facebook.com/majesticstudiosllc

Instagram - https://www.instagram.com/majesticstudios_jld/
User avatar
MarPabl
Platinum Member
Posts: 938
Joined: Sat Sep 24, 2011 12:09 am
Location: MX

Post by MarPabl »

We need further details to confirm if the upgrade can brick the Kronos. I tend to think that won't happen. That's the reason I talked about the possibility to restore previous OS and even directly downgrade without restoring.

I think you can even swap the SSD with the original OS and a new one with the fixes...
Current gear: :arrow: Access Virus TI2 Whiteout Keyboard (111/150), Access Virus TI2 Polar DarkStar Special Edition, Gibson Custom Lite 2013, Roland MV-8800 \:D/
1jordyzzz
Platinum Member
Posts: 688
Joined: Sat Mar 03, 2012 2:02 pm
Location: Indonesia

Post by 1jordyzzz »

danatkorg wrote: At Korg, we try not to promise things before we're ready to deliver them (though I'm sure that we slip up from time to time). That doesn't mean that we're not working on new stuff, though! We've already shipped the 1.5 update, which included several user-requested features, such as the doubling of user banks, quadrupling of sample & multisample slots, USB MIDI support, and additional RAM support. Many of the suggestions posted on this forum are on our wish lists as well. We are listening, and it is our goal to make instruments which are as useful to you as they possibly can be (and to us - we play them too!). We're grateful to have you as customers and fellow KRONOS-users, as this wonderful instrument continues to evolve.

Best regards,

Dan
Thanks Dan, i'm also grateful to have such fantastic instrument as the korg kronos.... and i'm sure many people here experience the same.
Love my kronos 88 :D
Love my yamaha psr s910 as well

Korg Kronos 88, Yamaha PSR s910, Korg C720, Yamaha DTX 520, Focusrite Scarlett 18i6, a pair of Yamaha HS80 in (soon not to be) an unproperly treated room..
billysynth1
Platinum Member
Posts: 1148
Joined: Sat Jan 22, 2005 5:09 am
Location: Australia/Melbourne

Post by billysynth1 »

My hat off to Heider. Prodigees do exist...look at the handful of piano prodigees, the rest of us are average or below average lol, we could never achieve such high level musicianship. Whats to suggest Heider is not a prodigee?

Billy
Yamaha C1 Grand Piano.
Korg Oasys 88, Jupiter 80
Kronos 88, V Synth GT
I am a student of classical piano...I am not a classical pianist.
Locked

Return to “Korg Kronos”