NDSP Quad Cortex

[mbenigni]

You do have a point there. Maybe design/purpose was there, just execution has been shitty.

#NotARant

I could drag us into a whole downward spiral of semantics here if you let me, man.

The launch was shit, famously so, and I'm glad I wasn't around for that. Nearly all the rest of the execution has been great: design, manufacture, distribution (notably, all the stuff that should have tripped up a software company)... first class. As much as everyone bemoans the absence of a "marketplace" (aka fools and their money LOL), the cloud infrastructure is extremely well implemented for purposes of distributing firmware updates, managing backups, and (finally) brokering plugin content. From where I'm standing it seems like NDSP has had the uncanny ability to get all of the really hard parts right, and then completely whiff when it comes to shit so easy they could have had their kids do it as a summer job (e.g. profile a few amps and throw us a bone; or even, send out a follow up email when you said you were going to send out a follow up email.) I can't even begin to explain it.

 

[Chocol8]

the cloud infrastructure is extremely well implemented

It’s like we speak entirely different languages that use exactly the same words, but the words have completely different meanings.

 

[mbenigni]

Doubt that’s much more than the UI. I am sure the actual modeling is done more directly on the DSP. If it was all at the software layer, it would be way less efficient, but OTOH, it would have been a breeze to port over the plugins. The struggle with plugins is a sure sign to me that they are writing for the DSP.

Never mind the technology. "Plugin compatibility" was a fool's errand when it was still an idea scribbled on a cocktail napkin. Trying to monetize a square peg when they already had a perfectly good round hole. They clearly were able to develop for the QC DSP at one point in time or the factory content wouldn't exist. And even if they'd lost those developers, they could have coasted on the capturing tech by just capturing and delivering content more frequently. I honestly thought that's what the whole "TINA" victory lap was about, but then they went into hiding again. (Which can only suggest that TINA didn't really work as well as predicted.)

 

[mbenigni]

It’s like we speak entirely different languages that use exactly the same words, but the words have completely different meanings.

I've never doubted it.

 

[Orvillain]

Doubt that’s much more than the UI. I am sure the actual modeling is done more directly on the DSP. If it was all at the software layer, it would be way less efficient, but OTOH, it would have been a breeze to port over the plugins. The struggle with plugins is a sure sign to me that they are writing for the DSP.

I honestly think it is just because they haven't had enough engineers for the last however many ... er.. years???

 

[AlbertA]

I read that QC runs on UNIX, I may be wrong. If so, the OS is there (whatever flavor it is, whatever that looks like)

Akshually - you mean Linux? I would be very surprised if it ran Unix.

 

[Mongillo19]

I honestly think it is just because they haven't had enough engineers for the last however many ... er.. years???

It's just strange to me even plugins have slowed down. Has it really been almost 2 years since a plugin launch? That doesn't seem right

So with that in mind.....what have they been doing?

 

[GTR37]

It's just strange to me even plugins have slowed down. Has it really been almost 2 years since a plugin launch? That doesn't seem right

So with that in mind.....what have they been doing?

Because you have to think any new plug ins they release
Going fwd would automatically be designed to fit the QC out of the gate
To release another plug in with the outdated code that they would again have to rework especially after knowing how much time it takes now would be absolutely bizarre

The other thing maybe they are working on another project like something that sits between NANO and QC

 

[Jarick]

I tried. I can't even look at Discord. The actual message-to-timestamps, usernames, etc. noise is just too much. It looks like a pile of slop to me.

View attachment 47913

I’ve never been on discord, is it like IRC?

 

[Jarick]

Seriously though... if NintendoDSP really existed.. we could have 'snake' on the touchscreen!!

The fucking marketplace would take 20 minutes to load though!

 

[LCW]

Akshually - you mean Linux? I would be very surprised if it ran Unix.

Probably Android lol

 

[mbenigni]

I’ve never been on discord, is it like IRC?

I've never been on IRC, is it like Discord?

 

[Los-D]

I could drag us into a whole downward spiral of semantics here if you let me, man.

The launch was shit, famously so, and I'm glad I wasn't around for that. Nearly all the rest of the execution has been great: design, manufacture, distribution (notably, all the stuff that should have tripped up a software company)... first class. As much as everyone bemoans the absence of a "marketplace" (aka fools and their money LOL), the cloud infrastructure is extremely well implemented for purposes of distributing firmware updates, managing backups, and (finally) brokering plugin content. From where I'm standing it seems like NDSP has had the uncanny ability to get all of the really hard parts right, and then completely whiff when it comes to shit so easy they could have had their kids do it as a summer job (e.g. profile a few amps and throw us a bone; or even, send out a follow up email when you said you were going to send out a follow up email.) I can't even begin to explain it.

#neverforget

 

[dws2384]

I don’t know if it true or not but I swear I had heard or read somewhere that they might have been thinking about selling the company like Darkglass / Korg
That would be a pretty big nightmare in the middle of PCOM if they lost some key devs

They act like a company planning to sell. Dont worry about the customer base as the buyers only look at the bottom line. Add No resources, release a new product to show new revenue streams. Doesnt sound that far fetched.

 

[Los-D]

They act like a company planning to sell. Dont worry about the customer base as the buyers only look at the bottom line. Add No resources, release a new product to show new revenue streams. Doesnt sound that far fetched.

 

[Chocol8]

I honestly think it is just because they haven't had enough engineers for the last however many ... er.. years???

Not enough of the right people with the right skills. You can't just throw a bunch generic software engineers at the problem, although plenty of management teams have tried that approach.

 

[mbenigni]

#neverforget

Hahahahaha. Oh my God. I think I actually had, somehow.

Touché.

 

[Chocol8]

They act like a company planning to sell. Dont worry about the customer base as the buyers only look at the bottom line. Add No resources, release a new product to show new revenue streams. Doesnt sound that far fetched.

I work with a lot of companies that do acquisitions, and they generally are smarter than that. Modern due diligence reports often include research on the company and any product brands perception which is really easy to mine online these days. Anyone buying NDSP would know about their reputation and client satisfaction issues.

A buyer that plans to take the QC hardware, rewrite or significantly update the software, and rebrand it might not care. Others will look at it and start chopping their offer price.

 

[Orvillain]

Not enough of the right people with the right skills. You can't just throw a bunch generic software engineers at the problem, although plenty of management teams have tried that approach.

It's not just that. People get moved from project to project, and swathes of domain knowledge aren't transferred properly due to ... ahem... spirited developers and lax middle management, so projects stall, regressions get inroduced, and all in all, shit can turn to shit shit faster than a proper vindaloo shit from shit heaven.

 

[laxu]

I don’t know if it true or not but I swear I had heard or read somewhere that they might have been thinking about selling the company like Darkglass / Korg
That would be a pretty big nightmare in the middle of PCOM if they lost some key devs

I fully expect this has been the plan all along.

Since the QC release, the Neural DSP head honchos have appeared in several Finnish business magazine articles over the years. I fully expect this is a push to get investors, increase stock value or aim for some larger company to buy them along the line. I don't fault NDSP leaders for doing this at all. This is a common pattern for many startups, and it's the type of thing that the C-level management should be doing.

I expect key devs would have shares in the company too, so the company being sold will mean money to them too. It doesn't necessarily mean that they will start leaving. People typically leave companies because they got a better offer elsewhere, or due to bad management/coworkers/working conditions.

If you look at Neural DSP's website Careers section, they don't have any specific job postings open but just open application. That to me means they feel they have enough people in key positions. It's not like you can just get any embedded DSP programmer off the street and expect them to be a productive addition to a team from day one.