Axe-Fx III Firmware 24.01 Release

SpaceGoD said:
I used to run an engineering department and you are, ideally, correct. However, I assume fractal does not have a bunch of qa engineers which leaves them with only a few options.

1 - put out a beta and have the early adopters find bugs, then integrate those changes. When you think you’ve gotten most of them put out a final release which a wider audience will then download and find more that are then fixed in incremental releases (ie 1.xx)
2 - keep the update in beta for an extended period of time and do a lot of in-house testing with whoever you can find. This takes away time from new features and updates will be a ton slower but a final release will likely be more stable.
3 - hire a bunch of people to qa it. Pay for new versions like other software.

The old adage holds that you can pick 2 of cheap, fast, or stable. With how easy it is to roll back, I, and I’m guessing most on here much prefer fast and free. If you’re concerned about stability just stay 1 whole version back, that’s what most people do in the corporate world with software.

Also as an aside, I’ve only had my axe for a month but I’m floored with not only the sound but also how Cliff manages to keep development rolling so quickly on such a massive feature set with relatively few bugs. Kudos to the whole team!
Click to expand...
Generally speaking if something is mission critical, you don't want to do major software upgrades immediately following the initial release with anything if you want bug free expectations (assuming your current version isn't having issues). You aren't really going to find that anywhere. Stuff sneaks through the cracks all the time whether we are talking music software, operating systems, or anything else and that's especially true the larger the application is. You can certainly minimize the risk, but it's always going to be there even when you are doing things right. If you require the Axe to be totally fail proof with the firmware, you should probably still be on a pre-23 version IMO.

That being said, I love this firmware update!!
 
I actually don’t have any issues with it. No pops on scene changes and I don’t use the tape echo or ping pong. Love that I am rediscovering the mk iv through the mark v model! It sounds great!
 
FractalAudio said:
My bad. I rushed the firmware out. Sorry I didn't meet your expectations.
Click to expand...
Working great for me, Cliff! Thanks for the all the work you put into adding so much value to our Fractal investment with these amazing updates! From one of the multitude of working musicians on here...THANK YOU and Happy Holidays!
 
SpaceGoD said:
I used to run an engineering department and you are, ideally, correct. However, I assume fractal does not have a bunch of qa engineers which leaves them with only a few options.

1 - put out a beta and have the early adopters find bugs, then integrate those changes. When you think you’ve gotten most of them put out a final release which a wider audience will then download and find more that are then fixed in incremental releases (ie 1.xx)
2 - keep the update in beta for an extended period of time and do a lot of in-house testing with whoever you can find. This takes away time from new features and updates will be a ton slower but a final release will likely be more stable.
3 - hire a bunch of people to qa it. Pay for new versions like other software.

The old adage holds that you can pick 2 of cheap, fast, or stable. With how easy it is to roll back, I, and I’m guessing most on here much prefer fast and free. If you’re concerned about stability just stay 1 whole version back, that’s what most people do in the corporate world with software.

Also as an aside, I’ve only had my axe for a month but I’m floored with not only the sound but also how Cliff manages to keep development rolling so quickly on such a massive feature set with relatively few bugs. Kudos to the whole team!
Click to expand...
I feel like the current process should be very effective.
Send a pre release beta to beta testers, Squash the low hanging fruit, post a beta in the forum for the user base to beta test, Fix any remaining bugs, Post an official release thats stable.

It's just recently the FW updates are rapid fire. Ever since gapless switching was added.
 
rcm78 said:
I feel like the current process should be very effective.
Send a pre release beta to beta testers, Squash the low hanging fruit, post a beta in the forum for the user base to beta test, Fix any remaining bugs, Post an official release thats stable.

It's just recently the FW updates are rapid fire. Ever since gapless switching was added.
Click to expand...
Yeah, gapless switching was very difficult and introduced a lot of special conditions that private beta testing didn't uncover.

Again, my bad. I rushed things out and we should've tested things more thoroughly. We will strive to do better in the future.
 
I think the debugging process with betas happens much more quickly when you are able to throw a lot more bodies at it, so the current system actually is the best approach in spite of the occasional hiccups. Nobody is forced to install beta firmware, so if it is a concern then don’t install it.

I would hazard a guess that a poll would reflect that most people are just fine with the current method.
 
FractalAudio said:
Yeah, gapless switching was very difficult and introduced a lot of special conditions that private beta testing didn't uncover.
Click to expand...
Still, much appreciate your continual chipping away at it and responding to reports to get it working across the board. 🙏

Beta testing (private & public) also seems a necessary part of QT/A given the sheer complexity (massive configuration space) of the software/hardware.
 
Anyone who’s been here for a while understands Fractal’s commitment to quality and constantly giving us more and more for the cash we’ve invested, how many companies do that these days?? I can list a dozen items around my house that could and should get firmware updates but never will again.

Stuff happens, we learn and move on, give Cliff a break man. Coding and testing are no simple tasks. I’ve been playing with rack (and lots of other tech) gear since the 80s and have never seen anything like the response and commitment we get from Fractal. This guy sprays out firmware like no one else and people take the time to complain, just blows my mind – but I am a glass-half-full kinda guy.

*edit, I've also been in IT for almost 30 years (servers/Unix/Linux). I've worked with many of the big vendors for a long time; dealing with patches, hot-fixes, firmware updates etc etc and again, have never seen the level of commitment that I see from Fractal.

If you have mission critical use for your stuff, use the stable firmware posted under the support section, if that doesn't function as promised, then complain.

I hope team Fractal gets some good down time until 2024. Thanks for all you do.

TAK
 
Last edited:
rcm78 said:
I feel like the current process should be very effective.
Send a pre release beta to beta testers, Squash the low hanging fruit, post a beta in the forum for the user base to beta test, Fix any remaining bugs, Post an official release thats stable.

It's just recently the FW updates are rapid fire. Ever since gapless switching was added.
Click to expand...
It generally is. The thing about the process is deciding when it's "stable". The beta group will never find all the bugs. Do you always test every scenario whenever there's a code change spending massive amounts of time doing so or do you do your best to catch the bugs people report, move on, and risk things falling through the cracks?

Not saying you're wrong for wanting something stable, I'm just trying to point out that fractal is trying to make a judgement call on speed of development vs stability and it doesn't always work out.
 
You must log in or register to reply here.
Back
Top Bottom