2020-09-30 00:13:15

The answer is yes. I would

2020-09-30 01:37:07 (edited by Boo15mario 2020-09-30 01:37:47)

I could get a monsters Razer laptop for $3000.edit, that might even have a desktop CPU and GPU.

2020-09-30 02:13:48

So Camlorn, I thought physics and engineering simulation softwarewas more resource intensive than programing. Or arcetect software for modeling buildings and such. But I would say, get a heavier machine, and don't buy Intel. Intel is still using a 14 nanometer process for their processors since 2014, and hasn't made any significant improvements other than raising core counts. Sure, Intel processors can hit 5 GHZ, but you need beastly cooling and liquid metal to sustain clocks like that. I think the eluktronics rp 17, though it is heavy, would be a good fit. It has an rtx 2060 GPU, so you aren't spending too much money on GPU. But you get a massively beastly AMD ryzen 4800H processor, with 8 cores and 16 threads. And you can chose no windows install, no ram and HDD, and order your own ram sticks, SSD and windows.

A learning experience is one of those things that say, "You know that thing you just did? Don't do that."

2020-09-30 02:21:20

From the specs I've read, that Lenovo recommendation earlier in the thread actually looks pretty sweet.

As far as the question about my own: I suspect it could handle a fair bit, but I don't ask a ton of it most of the time. I can tell you it doesn't lag when streaming video and running other tabs and stuff, but that's not saying a lot. It's also almost perfectly silent unless the ambient temperature is high, at which point I can hear its fan. Oh also, one time I put it on my pillow the wrong way and it sort of slid down between the pillow the my nightstand at a funky angle, which meant it got really hot. Nothing bad happened, but I take pains to make sure that doesn't happen anymore.

Check out my Manamon text walkthrough at the following link:
https://www.dropbox.com/s/z8ls3rc3f4mkb … n.txt?dl=1

2020-09-30 02:34:44

Jayde, you mention your system boots up in 17 seconds.A system with a really bad SSD and an optic drive can boot easily under 10 seconds, my  system has a good SSD, but its only sata, and it boots between 6-8 seconds. I bet a PCE  SSD can boot under 4 with no optic drive. The difference is nothing short of enormous. I highly recommend that if you can aford it, pick an SSD drive off Amazon,  The Samsung ones are the best in terms of speed and endurance if you can aford it, but you will never ever look back, or feel bad about the money spent on those drives. Here is another example. Installing windows 10 took 6 minutes, and about 4 of those was the slow flash drive copying the files to the pc.

A learning experience is one of those things that say, "You know that thing you just did? Don't do that."

2020-09-30 02:49:31

Got a surface laptop 3 last christmas that i love to death. Thing boots in like 5 seconds to NVDA, and apps install to fast lol.

2020-09-30 03:22:36

@enes
Yes and no.  Thing about that sort of software is that it's specialized.  These days, if you want to run a lot of that stuff, you do it all on the GPU and it basically doesn't care about what your CPU is.  That, or it's too resource intensive to run on a home computer (for example Dreamworks needs a massive supercomputer and it takes something like an hour per frame, if I recall--their particle simulator is open source).  Not saying there aren't exceptions, in fact I'm sure there's lots, but the modern world of scientific simulations isn't what you might expect, and it's not going to beat out compilers and build tools.

When I did my work on the Rust compiler, something like 3 years ago now, it brought a $4000 tower to its knees for about 30 to 45 minutes.  As in NVDA couldn't be used, go have lunch.  Fortunately for all involved we have incremental builds--it'll only recompile what changed--so that was maybe only twice a week.  The Rust compiler is an atypical project to work on, but as I said, work's codebase can take over a quad-core VM, and the current microservice I'm writing takes something like 2 or 3 minutes to build on the same VM if building from scratch (it's only on the order of a few thousand lines, but it brings in 300 or so dependencies after you account for dependencies of dependencies of..., which is a typical scenario as well).

My current workflow is this:

edit edit edit
cargo test
building for 20-30 seconds
running tests for 5-10 seconds
o it's still not working

repeat ad infinitum, and with the edits frequently taking under a minute each.

My Blog
Twitter: @ajhicks1992

2020-09-30 03:44:34

Camlorn, you mention quad core processors. How would the AMD processors, with massive core counts handle this type of workloads? For the rp17, reviewers say the 8 core processor is great for this sort of thing, especially since in that chassis they can all hit 4.2 GHZ.

A learning experience is one of those things that say, "You know that thing you just did? Don't do that."

2020-09-30 03:49:28

Enes, I take your point, and that's a good thing to know going forward. But there comes a point when improvement is not cost-effective. For instance, right now my drive boots fast enough for me. Jaws doesn't lag, even though it's a beast. Anything I use this computer for doesn't make it lag, except when I use IE on certain pages IE doesn't like (yes, I'm old-fashioned that way, but I'm moving gradually away from it). Other than that, this thing does exactly what I need it to. If I ever need to replace its drive, then sure I'd almost certainly upgrade. But until such time that that needs to occur, I'm not gonna bother.
You can buy a really good steak for twenty dollars (just work with rough numbers here). Or you can buy a really really good steak for forty dollars. Or an excellent steak for sixty dollars. But is the sixty-dollar steak really three times as good as the twenty-dollar steak? Will it even matter? My answer, at least in this analogy is no.

Check out my Manamon text walkthrough at the following link:
https://www.dropbox.com/s/z8ls3rc3f4mkb … n.txt?dl=1

2020-09-30 03:54:43

Probably they'd handle it fine.  My current tower is hex-core, used to be 8 core, it's generally a roughly linear speedup to a certain point.  Compilation is highly parallelizable, it'll just eat anything you throw at it until it's done compiling for the most part, at least with the bigger projects that have enough code for it.  People who, i.e., work on Chromium (several million lines and very much counting) will go for xeons with crazy core counts and server motherboards so that you can have 2+ sockets, sometimes.

If you're just starting out in programming though, a run-of-the-mill high-end laptop with 16GB of ram, an SSD, and a not-lame CPU will easily last years, after which you know whether or not you need to get the really crazy machines or not.  SO to be clear i'm not saying anyone thinking about programming should get multi-CPU xeon motherboards, indeed I don't have one myself.  You can make that decision if/when you need it later.  But it's not an unreasonable decision to make, if you wake up one day and are like "And now I compile Linux from scratch 20 times a day" or something, which are real jobs real people actually have.

My Blog
Twitter: @ajhicks1992

2020-09-30 06:58:37

@Simba
Thanks for that, I somehow missed it on apple's website.

2020-09-30 13:18:58

Jayde, the steak example  isn't accurate with SSDs. It isn't the point of deminishing returns. The improvement is so huge, so dramatic, that it is easily worth the price, like how a pc booting in 4 seconds will save you 13 seconds each boot, in addition to all programs, regardless of size loading instantaniously.

A learning experience is one of those things that say, "You know that thing you just did? Don't do that."

2020-09-30 14:54:10

@37 sort of has a point.  If you don't have an SSD and you get an SSD you probably won't ever want an HDD again.  Probably no need to upgrade the machine.  Even the most un-upgradeable laptops don't generally have hard drives that can't be swapped.  But it does generally speed up all of your day-to-day tasks at least some.  You'd be surprised how much stuff Firefox stores on disk for instance.

But I say sort of because if you're not doing something resource intensive and you are happy with what you have, shrug.  But if you have lag, it's one of the first big steps toward fixing it, generally.

My Blog
Twitter: @ajhicks1992

2020-09-30 23:45:04 (edited by defender 2020-09-30 23:45:40)

I like them because they are also much more durable, you can even ride in a vehicle while using your computer with no worries, which you should never do with a spinning HDD. When they start to fail naturally, instead of losing data, certain sectors just become read only.  Of course that's a problem after a while for programs and Windows it's self as they write to disk all the time, but it does mean that you can still get your data back, which is great.