FORUM RULES - READ BEFORE POSTING
We don't have a great many rules on this forum. The ones we do have are mostly common sense but there are a couple which you do need to pay close attention to. You can read the latest version of the rules here: viewtopic.php?t=5958

Problems with the Yamaha Genos registry memory

Yamaha Genos and Genos 2 digital workstation keyboard

Moderators: Derek, Buzzard, parametric, Saul

User avatar
Kaarlo von Freymann Finland
Member
Member
Posts: 36
Joined: Fri Jan 03, 2014 9:51 pm
10

Problems with the Yamaha Genos registry memory

Unread post by Kaarlo von Freymann »

Yamaha Genos
Yamaha Genos
Hi Everyone,

Having been a manufacturer of military target drones since 1968 I am well aware of the fact that manufacturers are reluctant to admit any defects in their products. Some of you may remember Ralph Nader. Of course GM would not admit the Corvair was unsafe.

https://www.coolridesonline.net/news-bl ... r-claimed/

You may be aware of the fact that Tyros 5/6 (it has served me well for many years) became what it was supposed to be only after 5 (!) :lol: updates.

As of this date the Genos has already in a few months time arrived at v. 1.3. To me that indicates that the manufacturer started selling it when the product was still unusable for gigging musicians. (I am an amateur and only gig to collect money for charities) I leave it those whose native tongue is English to find the right word for a company that believes selling unusable products is ok. In Japanese it is 常軌を逸した.

You may not be aware of the fact that trolling and Mafia tactics are used not only by Mr. Trump and Mr. Putin but also by large corporations like YAMAHA. There are many forums who make you believe they are not connected to YAMAHA, but after a while you realize they are. Which explains you are always told the product has no bugs, you the user are the problem. :/:

Which is not to claim i have not been genuinely helped with problems for which there is a solution, alas not a very intuitive one. (Having used keyboards since the YAMAHA 5700 I usually do find out the “how to” even when they forgot to explain it in the manual & reference manual. With Genos I was often lost)

But let there be no misunderstanding, if a software designer creates software that works even if you feel it should have been made differently, THAT IS NOT A BUG. For instance you can only save PRESET VOICES in FAVORITES, which is practically useless as you will want to use the excellent VOICE tweeking Genos incorporates and save the tweaked voice in favorites.

But if what you have memorized correctly - proof been able to call up - after a few days is corrupted, THAT IS A BUG.
I have the problem that registration bank buttons – admittedly complicated registrations – after a while often

- call up wrong tempo
- call up main A instead of INTRO
- call up VOLUME 127 for an instrument that was memorized at VOLUME 67 (earshattering so I dare not gig with Genos !)
- A SLIDER set to regulate STYLE VOLUME no more does that - it has reverted to regulating SONG VOLUME instead but still in the slider function
display says STYLE VOLUME. That is just one example of many SLIDER bugs.


Just to name another examples, If the function PARAMETER LOCK does not lock what you click to be locked, THAT IS A BUG.

Now we all know keyboards are mass-produced, and there is always the possibility that you happen to get a specimen that has a manufacturing defect. So in order to claim, the phenomenon is a BUG it must be present in specimens with a comparatively substantial difference in manufacturing date and/or serial number. The BUGs I believe to have are the same on 2 specimens.

So my question to you is, has anyone of you encountered the above described problems ? and if so, did you find a remedy ?
User avatar
Saul Ukraine
Site Admin
Site Admin
Posts: 16556
Joined: Sun Feb 01, 2004 12:00 am
20
Where Are You Located?: Nr Brighton, UK

Re: Problems with the Yamaha Genos registry memory

Unread post by Saul »

Well those are not just annoying bugs but actually impact quite substantially on the way you can use Genos. I would guess...and it is only a guess, that this is not a widespread issue? I would definitely be interested to hear of any more problems like this.

I'll fire off a message to Yamaha and see what they have to say.
Saul
Site Admin

If I have been able to help you in some way you can Buy Me A Coffee Here.

Image
Brand Ambassador For Dowina Guitars
Acoustic Guitar Reviews On Facebook
Subscribe To Our YouTube Channel!
Follow us on Twitter @YamahaMusicians
Follow Us On Facebook
Dowina Acoustic Guitars: Dowina Cabernet DC
Dowina Chardonnay GAC DS - Dowina Rustica GAC - Dowina Marus - Dowina Puella

LAG T88A, Breedlove Signature Concert Copper E, Breedlove Pursuit Exotic Concert CE Koa, Walden D640T, Tanglewood TRSJ-VS, Vintage VE2000GG.
User avatar
Gordon49 United States of America
Posts: 3
Joined: Thu Apr 26, 2018 2:35 am
5

Re: Problems with the Yamaha Genos registry memory

Unread post by Gordon49 »

I found a glitch with the Genos. On some of my registrations, the effect for the left voice turns off even if I save the registration with the left voice effect turned on. I was trying to use a compressor effect on the left hand bass, but some registrations turn it off. I contacted Yamaha, and they did get back to me. Supposedly this problem was supposed to be fixed with firmware 1.30, but they said apparently it wasn’t fixed completely so I guess there will be another firmware update down the road.
User avatar
Saul Ukraine
Site Admin
Site Admin
Posts: 16556
Joined: Sun Feb 01, 2004 12:00 am
20
Where Are You Located?: Nr Brighton, UK

Re: Problems with the Yamaha Genos registry memory

Unread post by Saul »

Hi Gordon, welcome to the forum :)

Thanks for letting us know about that. It's a shame that that these bugs have made it through the testing phase. Really it should not happen. Yamaha have a large r&d division and they do test over quite lengthy periods of time which makes it all the more disappointing that these things slip through but it does happen even if you have a large budget. With something as complex as Genos it is difficult to test for every scenario.

Software can be complex and the interaction with hardware often not straightforward.

It is incredibly frustrating but is also something we have learned to live with in the computer age. How many bugs does Windows have with every release? Even my beloved MacOS is not bug free.

I am sure though Yamaha WILL fix these things, they always do we just need to hang on in there. It will all come right in the end.
Saul
Site Admin

If I have been able to help you in some way you can Buy Me A Coffee Here.

Image
Brand Ambassador For Dowina Guitars
Acoustic Guitar Reviews On Facebook
Subscribe To Our YouTube Channel!
Follow us on Twitter @YamahaMusicians
Follow Us On Facebook
Dowina Acoustic Guitars: Dowina Cabernet DC
Dowina Chardonnay GAC DS - Dowina Rustica GAC - Dowina Marus - Dowina Puella

LAG T88A, Breedlove Signature Concert Copper E, Breedlove Pursuit Exotic Concert CE Koa, Walden D640T, Tanglewood TRSJ-VS, Vintage VE2000GG.
Davhol United States of America
Member
Member
Posts: 43
Joined: Mon Apr 16, 2018 12:44 am
5

Re: Problems with the Yamaha Genos registry memory

Unread post by Davhol »

Hmmm...sounds like we've paid 5 grand to beta test the Genos! Shouldn't it be the other way around???
User avatar
Kaarlo von Freymann Finland
Member
Member
Posts: 36
Joined: Fri Jan 03, 2014 9:51 pm
10

Re: Problems with the Yamaha Genos registry memory

Unread post by Kaarlo von Freymann »

Hi Gordon,
It sounds politically incorrect, but I am very happy others have problems just like I have! And I am also very happy I have finally found a site where the administrator(s) are not YAMAHA trolls in disguise. :D I had already given up hope that there still do exist people like Saul.

As i have posted (not only on this forum) I have consistent memory problems with Genos and YAMAHA Europe does not answer mails or phone calls. Fortunately the YAMAHA importer for Finland is very co-operative.(they are in business since decades and import all kinds of music-related products, not just YAMAHA) The gentleman responsible for the Genos was here for the third time and I demonstrated to him that the memory still even after update 1.2 and 1.3 is unreliable. Contrary to YAMAHA and many YAMAHA forums he does not subscribe to the tactic of claiming the problem is the user, not the keyboard. He said, "For someone like you wanting to use its full potential Genos still has too many bugs. Fortunately a lot of buyers just chose a style and a voice and play, and as we both know after 1.3 that works well. We will effectuate the agreed exchange of your Tyros 5/6 to Genos when YAMAHA has fixed the bugs we know are there. I am sorry to say, you will have to gig on the Tyros until then. :( Unfortunately it is impossible to estimate when the present bugs will have been fixed. Rewriting software is a time consuming job."

You will appreciate he has once again earned my highest respect.


Cheers

Kaarlo
User avatar
Saul Ukraine
Site Admin
Site Admin
Posts: 16556
Joined: Sun Feb 01, 2004 12:00 am
20
Where Are You Located?: Nr Brighton, UK

Re: Problems with the Yamaha Genos registry memory

Unread post by Saul »

I think he hit on an important point there Kaarlo. The reason more people are not mentioning the issues you mentioned is because as he says
a lot of buyers just chose a style and a voice and play
So I am assuming the issues are in fact present on ALL Genos keyboards and it is definitely not an isolated "bug".

Yamaha will know better than any one that their customers will often push their gear to the limits and beyond and if a feature is included then it should definitely work but of course this is the real world and software sometimes doesn't work the way you intend it to.

Look, I think Genos is a fantastic keyboard and for me personally I doubt I would have ever come up against most of the issues mentioned simply because my main instrument is acoustic guitar and I normally only skim the surface of what is available on any keyboard I have owned. However..If I am paying a very large chunk of change for something I do expect ALL features to work regardless of whether I use them or not. Perhaps an unrealistic expectation but it is one I can't seem to shake.

The best we can do here is air any issues that come to light and make sure that Yamaha is aware of them and hopefully is working on a fix.
Saul
Site Admin

If I have been able to help you in some way you can Buy Me A Coffee Here.

Image
Brand Ambassador For Dowina Guitars
Acoustic Guitar Reviews On Facebook
Subscribe To Our YouTube Channel!
Follow us on Twitter @YamahaMusicians
Follow Us On Facebook
Dowina Acoustic Guitars: Dowina Cabernet DC
Dowina Chardonnay GAC DS - Dowina Rustica GAC - Dowina Marus - Dowina Puella

LAG T88A, Breedlove Signature Concert Copper E, Breedlove Pursuit Exotic Concert CE Koa, Walden D640T, Tanglewood TRSJ-VS, Vintage VE2000GG.
User avatar
Derek Wales
Global Moderator
Global Moderator
Posts: 6499
Joined: Fri Dec 07, 2007 12:00 am
16
Where Are You Located?: Wales, UK

Re: Problems with the Yamaha Genos registry memory

Unread post by Derek »

As a software developer I will chip in and say that whilst you might expect software systems to be fault free it is very hard to achieve that, especially when things have to be built to time and budget to get to market. When it comes to real time software it is also extremely hard to test all input conditions and permutations of input values and expected results (i.e. obtaining full coverage testing). By way of example, the reason why safety related software is so expensive is the amount of additional design and testing effort required to get full coverage testing and prove it.

An industry standard metric is that under a normal test process, software will typically be released with a bug for every 1000 lines of source code. They need to be found and fixed (hopefully without introducing more).

So, a sad fact of life I am afraid. The thing that gets me mad is companies that do not have effective means of handling bug reports and I think this is where Yamaha and others could learn a thing or two from smaller developers. If a User can give me a nice clear use case of what leads to a bug that I can replicate then I stand a chance of fixing it. The worst you can get as a developer is a user giving you a huge diatribe about the software "not working" without giving any hints as to what they are doing. That is not going to help you find and fix problems. Most users of my software are pretty good, but you do get some....

Brad Robinson, author of Cantabile, is also an excellent example of somebody who is very responsive to bug reports and feature requests.

Yamaha by contrast, when you criticise the Montage and ask for a feature improvement will usually leap down your throat (or a certain rep of theirs will) and tell you that you "do not understand the Montage". I don't know what they are like with bug reports as I have not found any in the Montage. It is worth noting however that forums for major manufacturers are not usually the place to report bugs. It is better to call your local dealer and request technical support, but do be sure that you can find and articulate clear, reproducible examples of the problem if you can. :)
Regards
Derek Cook

http://www.carregddu.co.uk
http://www.echoes-music.co.uk
http://www.xfactory-librarians.co.uk
http://www.ex5tech.co.uk
User avatar
Kaarlo von Freymann Finland
Member
Member
Posts: 36
Joined: Fri Jan 03, 2014 9:51 pm
10

Re: Problems with the Yamaha Genos registry memory

Unread post by Kaarlo von Freymann »

Derek,
thanks for a posting with a huge amount of clarifying and understandable information. I never before read something like "As a software developer I will chip in and say that whilst you might expect software systems to be fault free it is very hard to achieve that ....An industry standard metric is that under a normal test process, software will typically be released with a bug for every 1000 lines of source code. They need to be found and fixed." Software is something we all work with every day but do not have the slightest understanding of except that

Photoshop must be immensely complicated considering what it can do in a split second to a 50 MByte picture.

That Google can come up with millions of results in a split second if you input "president John Kennedy" is surprising, but utterly unbelievable is it that it will come up with many pictures of me, an absolutely zero person immediately.

60 years ago during university holidays I worked as a test driver at NSU, AUDI, GM-Opel and BMW and I understood how the car worked and could with a big box of spare parts in the trunk get going again whether it was in the burning heat of the Sahara or - 35 C in Lapland. To day I am absolutely helpless if my 18 year old Range Rover has a problem, and worse, so is the Importer for Finland.

In the early seventies I built several Böhm and Wersi electronic organs. They had 16 feet IC:s one could solder, not 160 feet. The only thing I have done to Tyros and will do with the Genos is solder a resistor to both ends of the pitch bender pot to get about 1/4 note bend at full deflection because I used to play saxophone and 1/2 note is ok with a guitar but too much for a sax.

We all owe you very much for your excellent posting.

Cheers

Kaarlo
User avatar
Saul Ukraine
Site Admin
Site Admin
Posts: 16556
Joined: Sun Feb 01, 2004 12:00 am
20
Where Are You Located?: Nr Brighton, UK

Re: Problems with the Yamaha Genos registry memory

Unread post by Saul »

I totally accept that there is far more to engineering software than most of us are aware of. However if there is a "known" bug it should have been fixed. Genos is a premium keyboard, a flagship product for Yamaha. They need to be really on top of their game with this because people are investing a lot of money in it and putting their trust in Yamaha's reputation for quality and reliability. I KNOW they will fix these things but how long it takes them to do so is the key thing here.

Reputations take years to build and only moments to destroy. Genos is not the only game in town. The KORG PA4x is a great keyboard and at £2,814 new is a lot cheaper than Genos, which at the lowest prices I have seen (£3,443) is over £600 more. Is it a better keyboard? I can't say as I am not experienced enough with arranger keyboards to give an opinion. On gut instinct I would still go for a Genos but that's just me.

I do think it would take a LOT for any Yamaha Tyros/Genos owner to migrate to KORG ;)
Saul
Site Admin

If I have been able to help you in some way you can Buy Me A Coffee Here.

Image
Brand Ambassador For Dowina Guitars
Acoustic Guitar Reviews On Facebook
Subscribe To Our YouTube Channel!
Follow us on Twitter @YamahaMusicians
Follow Us On Facebook
Dowina Acoustic Guitars: Dowina Cabernet DC
Dowina Chardonnay GAC DS - Dowina Rustica GAC - Dowina Marus - Dowina Puella

LAG T88A, Breedlove Signature Concert Copper E, Breedlove Pursuit Exotic Concert CE Koa, Walden D640T, Tanglewood TRSJ-VS, Vintage VE2000GG.
User avatar
Derek Wales
Global Moderator
Global Moderator
Posts: 6499
Joined: Fri Dec 07, 2007 12:00 am
16
Where Are You Located?: Wales, UK

Re: Problems with the Yamaha Genos registry memory

Unread post by Derek »

Agreed. if it's known and replicable then it should be fixed, but another issue to consider is prioritisation, based on:
  • How often does it happen (actually more like how often are users likely to hit the trigger conditions that lead to a failure. Software is systematic and not random in how it fails, although it might seem that way. Failure modes in software are as a result of trigger conditions causing a code branch to a piece of software with a problem in it (which has not been fully tested). If you don't hit the conditions for the bug then failure probability for that bug is zero. If you hit the trigger conditions for a bug then failure probability is 100%
  • What impact does it cause. E.g. data loss or mild annoyance?
  • How many users are likely to be affected (depends on how many people have the same workflow in which the bug manifests - E.g. I never suffered from the EX5 sequencer issues as I NEVER used its sequencer)
  • Can it be replicated?
  • Is it easy to fix?
  • Is there one or more workarounds?
At the end of the day, even large manufacturers will have finite resources, so will prioritise. If one user reports an obscure hard to replicate bug which is annoying or 100 users report a replicable bug that causes data loss, which one is going to get the priority?

On the latter point one drawback of Yamaha still using custom hardware is that if the "bug" has manifested itself in the programming of (potentially mask programmed) chips, then it will be very hard to fix. That's where Korg's Kronos COTS hardware approach is much more flexible as evidenced by the number of feature upgrades they have done over its life.
Regards
Derek Cook

http://www.carregddu.co.uk
http://www.echoes-music.co.uk
http://www.xfactory-librarians.co.uk
http://www.ex5tech.co.uk
User avatar
Derek Wales
Global Moderator
Global Moderator
Posts: 6499
Joined: Fri Dec 07, 2007 12:00 am
16
Where Are You Located?: Wales, UK

Re: Problems with the Yamaha Genos registry memory

Unread post by Derek »

Kaarlo von Freymann wrote: Fri Apr 27, 2018 1:20 am Derek,
thanks for a posting with a huge amount of clarifying and understandable information. I never before read something like "As a software developer I will chip in and say that whilst you might expect software systems to be fault free it is very hard to achieve that ....An industry standard metric is that under a normal test process, software will typically be released with a bug for every 1000 lines of source code. They need to be found and fixed." Software is something we all work with every day but do not have the slightest understanding of except that

Photoshop must be immensely complicated considering what it can do in a split second to a 50 MByte picture.

That Google can come up with millions of results in a split second if you input "president John Kennedy" is surprising, but utterly unbelievable is it that it will come up with many pictures of me, an absolutely zero person immediately.

60 years ago during university holidays I worked as a test driver at NSU, AUDI, GM-Opel and BMW and I understood how the car worked and could with a big box of spare parts in the trunk get going again whether it was in the burning heat of the Sahara or - 35 C in Lapland. To day I am absolutely helpless if my 18 year old Range Rover has a problem, and worse, so is the Importer for Finland.

In the early seventies I built several Böhm and Wersi electronic organs. They had 16 feet IC:s one could solder, not 160 feet. The only thing I have done to Tyros and will do with the Genos is solder a resistor to both ends of the pitch bender pot to get about 1/4 note bend at full deflection because I used to play saxophone and 1/2 note is ok with a guitar but too much for a sax.

We all owe you very much for your excellent posting.

Cheers

Kaarlo
Glad that you found it informative. I've done software engineering from hand optimised assembler code (in the days when it mattered for performance - heck I even started in the day when you had to hand assemble your program and manually type in the processor op codes and operands!) through to safety critical software systems written in SPARK Ada (the "safest" software language going), where you have to get 100% code coverage during testing or justify why you can't and what the risks are.

Brilliant example:
  • It is easy to test a divide by zero exception handler in a test rig, simply run a test case where you divide by zero and ensure the handler runs and produces the right outcome (logging the error and/or halting the processor)
  • testing a processor overheating exception handler is a little harder, unless you get the blow torch out...... (potentially on a board costing 10's of thousands of pounds.. ) ;)
  • Assuming the same exception handler is used for both, you test "by inference". You make an argument that if the handler works for divide by zero then it will work for processor overheating (assuming the hardware is working)
Like you, I've also got a hardware background (in fact, I find the best software engineers do as an appreciation of the hardware helps). Worked my way up from transistors, TTL logic, PLDs, FPGAs.....

That's interesting what you did for the Genos pitch bend. I though it was odd first, but then realised that the increment you can set on most synths is in semi-tone steps. :)
Regards
Derek Cook

http://www.carregddu.co.uk
http://www.echoes-music.co.uk
http://www.xfactory-librarians.co.uk
http://www.ex5tech.co.uk
User avatar
Kaarlo von Freymann Finland
Member
Member
Posts: 36
Joined: Fri Jan 03, 2014 9:51 pm
10

Re: Problems with the Yamaha Genos registry memory

Unread post by Kaarlo von Freymann »

Dear Derek,

now you definitely broke the sound barrier - or perhaps better put left with a s"woooosh" to spheres I am not able to follow you to. Its a problem I run into in articles about the universe. I can understand that a "light year" is a distance unit like mile or km, and not light like some whiskies or a nice year, but "a black hole" is beyond what I can grasp. As the Genos is black which may be cheaper than the finish of the Tyros and definitely shows scratches and wear and dust more, just like the YAMAHA 5700 I still have (for sentimental reasons.) Maybe its memory has "black holes" 8O

Alas your only mistake is, you overestimated my ability to understand. Underestimating a foe is a terrible mistake, overestimating a friend is a compliment.

- hand optimised assembler code
- exception handler
- manually type in the processor op codes and operands!
- assuming the same exception handler is used for both, you test "by inference"
- where you have to get 100% code coverage during testing or justify why you can't and what the risks are.
:? :roll:

But I believe I understood what you wanted to say though I did not understand what you said.

What you wrote on Fri Apr 27, 2018 9:56 pm was an excellent analysis in which I was able to follow the reasoning and and very much agree. THANKS.

Cheers

Kaarlo
User avatar
Derek Wales
Global Moderator
Global Moderator
Posts: 6499
Joined: Fri Dec 07, 2007 12:00 am
16
Where Are You Located?: Wales, UK

Re: Problems with the Yamaha Genos registry memory

Unread post by Derek »

Fair point. :D If you get the gist then that is good. If you are interested and want me to explain the terms, I am happy to do so.

Funnily enough I was thinking this morning of another analogy to explain the challenge of projects which is less technical :)

As a project manager, one of the things you are taught about is the Time/Cost/Quality Triangle

Image

And this is a great way of showing people the constraints that a project (like developing a new synth/workstation has). The idea of this triangle is to illustrate that a project will always have these three fundamental constraints:
  • Time - How long have you got to do the work? Often this is constrained by "time to market" and getting there before your competitors do. But you r will also be constrained by the development schedule, how many engineers you have to do the work....
  • Cost - What is your budget for development? This will be factored by your market analysis? How big is the market? How much are people prepared to pay for the product? What is the profit margin? What is the R&D budget from the margin that is allocated to YOUR project? And during the project you will be monitoring your expenditure against budget if you go over budget on one area you may need to cut back on others (or go over budget and affect profit)
  • Quality - are you targeting high end or low end in terms of how good the product is? Is it a product that is single use (e.g. a cotton bud) or a high end synth?
The triangle is showing you that a perfectly executed project should hit the centre spot of the triangle where time/cost/quality are optimised.

It shows you that, say, you want for a bug free system (higher quality) then this means you must spend more time on getting your project requirements right and spend more time in testing and fixing. Thus if you pull over on the quality axis on its own you will impact cost and time (it will cost more and take longer). If you want high quality and keep the duration the same, it will massively impact cost as you will need more resource. You get the idea?

I know this may seem to have gone way off topic (as we sometimes do here :) ) but in the context of the original debate, and whilst I am in no way defending Yamaha, I just thought my experience in delivering complex projects might give people an appreciation of the constraints that companies are under, and why we live in a world where bugs do get out into the real world.

As an engineer on the civil service I was a technical perfectionist (hard over on that quality axis). As a project manager in privatised company (spun off from the civil service dept I started in), then you need a dose of realism about time and budget as well. :)

I have two other thoughts for you. Our systems are getting increasingly complex. Compare your Genos to maybe the first synth with a microprocessor in, the Prophet 5. That probably had a few 1,000 lines of code in (so on my metric above, maybe a few bugs in it). Consider Microsoft Windows (yikes), which is generally thought to have over 50 millions line of code in it. A codebase that size unleashed into the world will have around 50,000 bugs in it. Probably less as it will be based on old mature code as well, but as new features are added then bugs will be introduced.

The other one is on effective testing and getting users involved as quickly as possible, because users will break systems in ways that engineers will never envisage just because the will approach the systm in a different way. The more complex the system (with different ways of doing things) and the more diverse a user base, then you will pick up more bugs than one test engineer sitting in a room always running through his test points the same way.

Hope that is an interesting insight.

As I said, and have hopefully shown, it is not realistic to expect bug free systems. The thing that makes a difference is the approach, openness and honesty that a company has in supporting the fixing of reported bugs. That is where Yamaha can "become very corporate" as amply demonstrated on the EX5 bugs.
Regards
Derek Cook

http://www.carregddu.co.uk
http://www.echoes-music.co.uk
http://www.xfactory-librarians.co.uk
http://www.ex5tech.co.uk
User avatar
Kaarlo von Freymann Finland
Member
Member
Posts: 36
Joined: Fri Jan 03, 2014 9:51 pm
10

Re: Problems with the Yamaha Genos registry memory

Unread post by Kaarlo von Freymann »

Derek wrote: Sat Apr 28, 2018 9:02 am Fair point. :D If you get the gist then that is good. If you are interested and want me to explain the terms, I am happy to do so.

Funnily enough I was thinking this morning of another analogy to explain the challenge of projects which is less technical :)

As a project manager, one of the things you are taught about is the Time/Cost/Quality Triangle

Image

And this is a great way of showing people the constraints that a project (like developing a new synth/workstation has). The idea of this triangle is to illustrate that a project will always have these three fundamental constraints:
  • Time - How long have you got to do the work? Often this is constrained by "time to market" and getting there before your competitors do. But you r will also be constrained by the development schedule, how many engineers you have to do the work....
  • Cost - What is your budget for development? This will be factored by your market analysis? How big is the market? How much are people prepared to pay for the product? What is the profit margin? What is the R&D budget from the margin that is allocated to YOUR project? And during the project you will be monitoring your expenditure against budget if you go over budget on one area you may need to cut back on others (or go over budget and affect profit)
    [*]Quality - are you targeting high end or low end in terms of how good the product is? Is it a product that is single use (e.g. a cotton bud) or a high end synth?


The triangle is showing you that a perfectly executed project should hit the centre spot of the triangle where time/cost/quality are optimised.

It shows you that, say, you want for a bug free system (higher quality) then this means you must spend more time on getting your project requirements right and spend more time in testing and fixing. Thus if you pull over on the quality axis on its own you will impact cost and time (it will cost more and take longer). If you want high quality and keep the duration the same, it will massively impact cost as you will need more resource. You get the idea?


I know this may seem to have gone way off topic (as we sometimes do here :) ) NOT AT ALL ! but in the context of the original debate, and whilst I am in no way defending Yamaha, I just thought my experience in delivering complex projects might give people an appreciation of the constraints that companies are under, and why we live in a world where bugs do get out into the real world.

As an engineer on the civil service I was a technical perfectionist (hard over on that quality axis). As a project manager in privatised company (spun off from the civil service dept I started in), then you need a dose of realism about time and budget as well. :)

I have two other thoughts for you. Our systems are getting increasingly complex. Compare your Genos to maybe the first synth with a microprocessor in, the Prophet 5. That probably had a few 1,000 lines of code in (so on my metric above, maybe a few bugs in it). Consider Microsoft Windows (yikes), which is generally thought to have over 50 millions line of code in it. A codebase that size unleashed into the world will have around 50,000 bugs in it. Probably less as it will be based on old mature code as well, but as new features are added then bugs will be introduced.

The other one is on effective testing and getting users involved as quickly as possible, because users will break systems in ways that engineers will never envisage just because the will approach the systm in a different way. The more complex the system (with different ways of doing things) and the more diverse a user base, then you will pick up more bugs than one test engineer sitting in a room always running through his test points the same way. ((i))

Hope that is an interesting insight.

As I said, and have hopefully shown, it is not realistic to expect bug free systems. The thing that makes a difference is the approach, openness and honesty that a company has in supporting the fixing of reported bugs. That is where Yamaha can "become very corporate" as amply demonstrated on the EX5 bugs.
I wrote a THANKs text, and saved it, but it is unreadable, will be back.
Post Reply