Support and discussion for all of Akai’s modern standalone MPCs including the MPC X / X SE, MPC Live 1, 2 & 3, MPC One / One+, MPC Key 37/61.

MPC 2.8 worth the wait?

38
78%
3
6%
8
16%
By Lumi Sat May 23, 2020 1:24 pm
NearTao wrote:Image

Still... 2.8.1 will hopefully address some of these imperfections :D

Yeah actually the touchscreen/qlinks optimizations and new midi protocol are a huge step ahead.
But it probably also has side effects and all architecture needs to be fixed again, so now Akai duty should be to make the other half of the job.
And if they do it, we have a beast.
By Eyalc Sat May 23, 2020 1:52 pm
They need to do two things or else this is going to happen constantly:

1. Fix the foundation. When you add things on a shaky foundation, this is the result.

2. Expand the testing coverage. No way something like this shouldn't have been caught in a beta test. It's almost like all of the testers have the exact same workflow and usage patterns. If so, they need to expand.
User avatar
By NearTao Sat May 23, 2020 2:26 pm
Eyalc wrote:They need to do two things or else this is going to happen constantly:

1. Fix the foundation. When you add things on a shaky foundation, this is the result.

2. Expand the testing coverage. No way something like this shouldn't have been caught in a beta test. It's almost like all of the testers have the exact same workflow and usage patterns. If so, they need to expand.


Personally, I'm less concerned that a regression happened... and more concerned that it doesn't seem to have been acknowledged in a meaningful way, and Akai is not really good at explaining what they will do to make sure the regression doesn't happen again.

but that's my two cents.
By Lumi Sat May 23, 2020 2:55 pm
Eyalc wrote:They need to do two things or else this is going to happen constantly:

1. Fix the foundation. When you add things on a shaky foundation, this is the result.

2. Expand the testing coverage. No way something like this shouldn't have been caught in a beta test. It's almost like all of the testers have the exact same workflow and usage patterns. If so, they need to expand.

I think beta testing is not working so good.
I mean, there are some issues that a good beta tester should notice in a minute, like the conflict between the qlink overlay and the data dial.
Maybe they didn't have time to test all? They were focused on checking multimidi?
I don't know but before releasing an update, you first have to check if all that worked is still working as before.
Take a look at getsatisfaction forum, there is a huge list of people having problems with 2.8 and mrs. Eliza is not replying anymore.
By Eyalc Sat May 23, 2020 2:58 pm
NearTao wrote:
Eyalc wrote:They need to do two things or else this is going to happen constantly:

1. Fix the foundation. When you add things on a shaky foundation, this is the result.

2. Expand the testing coverage. No way something like this shouldn't have been caught in a beta test. It's almost like all of the testers have the exact same workflow and usage patterns. If so, they need to expand.


Personally, I'm less concerned that a regression happened... and more concerned that it doesn't seem to have been acknowledged in a meaningful way, and Akai is not really good at explaining what they will do to make sure the regression doesn't happen again.

but that's my two cents.


Great perspective. And I agree 100% about the lack of acknowledgement. I think Akai has been in "fire and hope" mode for a while now. Which is often accompanied by lack of communication.

Always more costly to fix foundation stuff when there's shiny stuff sitting on top of it. And I think that's where they are. Pull here, that breaks there. Pull there, that breaks here. And a lack of communication shows that they very well may not know what will break next.

Given that, I don't think they're in an incredibly bad spot. The foundation just needs attention.
By Eyalc Sat May 23, 2020 3:03 pm
Lumi wrote:
Eyalc wrote:They need to do two things or else this is going to happen constantly:

1. Fix the foundation. When you add things on a shaky foundation, this is the result.

2. Expand the testing coverage. No way something like this shouldn't have been caught in a beta test. It's almost like all of the testers have the exact same workflow and usage patterns. If so, they need to expand.

I think beta testing is not working so good.
I mean, there are some issues that a good beta tester should notice in a minute, like the conflict between the qlink overlay and the data dial.
Maybe they didn't have time to test all? They were focused on checking multimidi?
I don't know but before releasing an update, you first have to check if all that worked is still working as before.
Take a look at getsatisfaction forum, there is a huge list of people having problems with 2.8 and mrs. Eliza is not replying anymore.


Yeah man, I agree 100%. Could very well be that they didn't have time. With a solid architectural foundation, there are some things that you shouldn't have to expect to test, because you have a clear map of what components affect what other components. When you have surprises like this, it reflects a foundation that isn't tight, and they didn't know that "this affects that". I get it, it happens in software. Not being overly critical. Just sharing my perspectives. This is why I keep stressing - they need to address the foundation and stop adding stuff on top.

And IMO, always worse when there are issues and the vendor disappears. Never a good look.
By Lumi Sat May 23, 2020 3:19 pm
@ Eyalc

Yeah exactly. Quite scary ahah.
I am sure that the new midi protocol is a big changing to all the architecture and I think this is why it took so much time.
And of course all is related so any changing or new feature means that you have to refix all the foundations.
I just wonder if they checked this before releasing the update or just made all in a rush to put it on Live2 the day of its release.
And I hope they will now make a 2.8.1 to fix all and we are all happy.
By Eyalc Sat May 23, 2020 3:32 pm
Lumi wrote:I just wonder if they checked this before releasing the update or just made all in a rush to put it on Live2 the day of its release.


I think this is probably right.

But I don't see why though. There's nothing on the Live 2 that make these features more desirable / required.

But, such is Akai.
By Lumi Sat May 23, 2020 3:40 pm
Eyalc wrote:
Lumi wrote:I just wonder if they checked this before releasing the update or just made all in a rush to put it on Live2 the day of its release.


I think this is probably right.

But I don't see why though. There's nothing on the Live 2 that make these features more desirable / required.

But, such is Akai.

Well that's marketing.
A new Mpc with a new firmware!
It sounds good, isn't it?
User avatar
By Lampdog Sat May 23, 2020 3:53 pm
Eyalc wrote:They need to do two things or else this is going to happen constantly:

1. Fix the foundation. When you add things on a shaky foundation, this is the result.

2. Expand the testing coverage. No way something like this shouldn't have been caught in a beta test. It's almost like all of the testers have the exact same workflow and usage patterns. If so, they need to expand.

There should be a test sheet with PRECISE functions and steps to check off.
User avatar
By Monotremata Sat May 23, 2020 4:51 pm
Lampdog wrote:There should be a test sheet with PRECISE functions and steps to check off.


When Ive done internal beta testing, we were always given scripts to use to attempt to break things and test functionality. The public beta testing Ive done, there wasn't anything to it. I remember both Wave Arts and FXPansion just gave out apps/plugins and said 'Here you go let us know if you have any problems' but nothing specific to look at or try out. If you used a limited feature set in one that worked for you, you may never encounter bugs so they'll go completely unreported.
By Eyalc Sat May 23, 2020 10:32 pm
Lampdog wrote:
Eyalc wrote:They need to do two things or else this is going to happen constantly:

1. Fix the foundation. When you add things on a shaky foundation, this is the result.

2. Expand the testing coverage. No way something like this shouldn't have been caught in a beta test. It's almost like all of the testers have the exact same workflow and usage patterns. If so, they need to expand.

There should be a test sheet with PRECISE functions and steps to check off.


YES.
By Eyalc Sat May 23, 2020 10:33 pm
Monotremata wrote:
Lampdog wrote:There should be a test sheet with PRECISE functions and steps to check off.


When Ive done internal beta testing, we were always given scripts to use to attempt to break things and test functionality. The public beta testing Ive done, there wasn't anything to it. I remember both Wave Arts and FXPansion just gave out apps/plugins and said 'Here you go let us know if you have any problems' but nothing specific to look at or try out. If you used a limited feature set in one that worked for you, you may never encounter bugs so they'll go completely unreported.


This is what I think the problem is.
User avatar
By Monotremata Sun May 24, 2020 12:57 am
Eyalc wrote:This is what I think the problem is.


Especially with the MPC. There are so many ways to get what you need done. Its the king of 'workarounds', you might be pulling something off with another function and not touching half of the features inside the OS. Thats alot of shit to skip over, and the way the code itself interacts with the rest of it is kinda important. One teeny tiny change to fix one feature everyone else uses, might break a feature that only 10 people use, so those bugs will just get swept under the rug if those 10 people aren't involved in the beta test.