Posts by Cardinal Fang

Why not get a NUC ?

The problem is that if you had to support ARM Linux there weren't really better options.

Yes, you can build with a cross compiler which works but is hard to maintain, test, emulator, well sort of but you rely on the emulator being accurate (and they weren't for ARM).

Mac wasn't really a good option apart from using them for OS/X support. Accuracy was an issue if nothing else.

Buy one of the server boxes ?. Well we tried that, they were relatively expensive, hard to source and died young. And the larger ones were really expensive and very hard to source.

So that left a 4G Pi. Which worked mostly, slower than we wanted because we didn't typically ship until all platforms built and passed FVT/performance. (Burnt a few times by failures on one platform being ignored but actually warnings of failures that would bite someone on one of the other platforms eventually).

So, the 16G Pi would be a godsend in that situation, in practice it'd be a lot faster, not just CPU speed but lack of memory pressure.

It's fairly niche admitted but faster is faster and it's actually ARM so testing was mostly an accurate indicator of what would happen when scaled up.

Other advantages

I was able to call 'C' more or less directly from Tcl (man swig) and for wrapping test code I could get performance 1000x better than Perl could and write and deploy test case code about 100x faster than I could do the same in C.

The developers hated me, because the sheer rate that I could blast requests at the servers I'd been asked to stress test meant I uncovered all sorts of weaknesses that would otherwise only have shown up when the code being tested was deployed by large customers.

The payoff - :). I was never asked to play Red Team again ....

The catch is

The database has to be more reliable than whatever it's replacing.

Fairly hard to beat direct memory addressing in the kernel or a filesystem on disk.

I'm not saying the idea is bad but it's been tried before and not being robust in the face of [memory | disk] errors has been the problem.

But Snaps were so SLOOOOOOW

In the end I pulled everything Snap based so my machine wouldn't be sitting there for hours waiting for updates to complete.

The only pain I feel is from having to update firefox by hand now and then, the cumulative pain from that since I consigned snap to the garbage bin of history would still be less than that of one snap update.

Wasn't bandwidth from the larger packages AFAIK, I play games on Steam and multi-GB updates there are quite speedy.

Seriously, if you want something to succeed it has to not kill performance.

You can cross 'enterprise use' off the application areas for that reason alone, people do not like coming into work, turning the laptop on then having to wait hours before the day can start.

I have no idea why it was so slow in my case, not my problem beyond finding a cure.

Note: I actually like Ubuntu otherwise.

Marketting is all

Apple is already doing well out of ARM laptops.

Performance isn't wonderful, but better app performance and compilers (and lots of marketing :) ) make up for a lot. Apple have a big edge being able to tune their software and hardware together for what's almost a single configuration.

You can tell actual performance is poor from one single area, games.

As with some examples above, my years old gaming rig blows modern Apple laptops away for performance.

Even Intel laptops still have an edge over Apple on raw performance but are held back by that boat anchor called Windows. Install Ubuntu and generally life is a lot better - alas for Intel, by default their hardware generally ships with Windows.

Catch is

1 of 3 bugs were memory management.

So Rust doesn't necessarily get rid of the other 2/3, and may even introduce some new bugs.

I'd be prepared to be that had memory check tools been available when sudo was written it would have been a lot less problematic, probably not memory safe but a lot less grief than it's been.

I'm not saying Rust is a bad language, and I'd call sudo ideal for Rust. The performance hit for something like sudo is well worth the cost, but assuming changing to a memory safe language is going to solve ALL your woes is being very naive. It's possible to do stupid in any language and in particular unexpected privilege escalation logic bugs is just as likely in Rust as in C.

BTDT

Usual problems, it won't work under difficult conditions.

There have been multiple 'solutions' to improve Ethernet throughput, they all fall apart when they have to work with bad latency and dropped packets.

One of the GPL conditions is

You can't impose extra conditions on users on top of the GPL'd code.

IBM is screwed here.

I'm not blaming RedHat here, they walked that very fine line very well. This is a toe over.

I'm an ex IBM employee, one of my internal comments was that if I hadn't already retired I'd have quit over this stupidity, some of that code was mine. (A very small part).

Why it's stupid, if it ends up in court IBM loses either way. They either lose the right to GPL code or if they win someone more feral will come along and rip the carpet from under them.

Domestic use

Lets see, 65" TV, my old gaming box does fine as a set top box hidden behind that.

That was the problem with NUC's, there weren't that many uses where a cheap PC (in my case free) wouldn't do the job so pushing them at a premium price wasn't going to work. I'm sure there were some use cases but when push came to shove they weren't compelling enough on anything but size and there a Pi was even smaller.