We probably need to sit down and come up with some tests to decide whether hardware is going to be fast enough or not. I’m not convinced just looking at the number of cores etc will be enough, we know what effect sticking an SSD in a machine does to it’s responsiveness.
Bonus points if the tool could point out if things like increasing RAM, switching to an SSD, etc would likely make it worth saving borderline cases.
Guess check 1 is “Is this hardware 64-bit?”
(This is a longer discussion, so I’ve moved the posts to a new topic.)
Yeah, we’re going to need to figure out some benchmarks. real-world feel is the key. Some thoughts:
I’m not sure what other “real world” things we could do to assess the “feel” of using a laptop…
Yeah, I suspect boot/load time are reasonable metrics here and are a better measure of performance than “I have a X core machine running at Y GHz”. I guess we’ll need to find someone non-technical to get a perspective of what the average person would find acceptable or not to provide some form of go/no-go metrics here. (Which opens up the fact that what’s acceptable to one person might not be for another.)
I know that @FionaMcG’s son struggled to use the rather ancient laptop that I’d lent her to play with for some school work, which seems to be because it wasn’t powerful enough, so it would be good to understand that in some better detail. May be running Speedometer 3.1 ?
How much we can run simultaneously is going to be in large part a facet of how much RAM the machine has. It would be interesting to get an idea of how many things the remake staff typically have open on their PC at the same time, I suspect any group I could poll might be heavily skewed towards “power users” with eleventy billion browser tabs open. Though open to finding that it’s closer to the norm than I’d thought…
100% with hardware tests.
One thing that’s going to be tricky is that ideally we want to be running with Linux installed on the devices storage rather than via USB as that’s going to heavily skew results, though ideally we’d also be able to perform the tests before commiting to blowing away what the owner has installed on the laptop. Maybe we’ll need to start capturing some metrics from a USB boot, then also capture some more real world measurements once installed and over time build up a picture of how the two are related.