Fork me on GitHub
#off-topic
<
2018-07-26
>
vemv12:07:15

Anyone got the 6-core MBP? I'd be interested in how in performs under a 'developer' workload (as opposed to rendering stuff or similar benchmarks), sustained over e.g. 8 hours. Something fairly typical for a developer would be to have running 1 VM, 1 IDE, Chrome, Slack, a web server, a database, some auxiliary processes. Not too demanding, but enough to keep the 6 cores entertained. The big question for me is - will it overheat? (I'm worried even after >) The thing I love about my MBAir is that it never overheats so I can actually have it on my lap. Even if I never had the MBP on my lap, I'd find the heat / fan noise unsettling

borkdude12:07:33

I read somewhere it’s more likely that it performs worse than previous MBPs under the same load, to maintain temperature

4
bherrmann12:07:16

I would think, unless you are looking for prime numbers or transcoding video.... it is probably fine.... You might benchmark your current system, then obtain the new one (14day no question return policy) , and verify that it works better for you and your workload

vemv12:07:56

Good call on the 14d policy!

henrik12:07:01

The new MB had thermal throttling problems, but those are fixed now. https://www.macrumors.com/2018/07/24/throttling-fix-2018-macbook-pro-improvements/

borkdude12:07:19

@henrik ah, then it’s fixed

borkdude12:07:01

I’m curious about the keyboard as well. Is the keyboard on the previous (new) MBP the same as the one you buy as bluetooth?

borkdude12:07:17

I’ve never had problems with the bluetooth keyboard

vemv12:07:23

Magic Keboard 1 or 2? I use the 1, I would still use it for new MBP because it moves/clicks more

vemv12:07:30

So, new MBP has a new iteration of the 2 tech, more silent and debris-resistant

borkdude12:07:52

ok, but the current MK 2 I have is the same model as were in the previous MBPs?

vemv12:07:14

I would say so, same kind of key

vemv12:07:41

Yeah I heard of the throttling issues but I'm worried even after those. It'd kinda common to see MBPs that burn your hands. But it can be hard to determine if Joe Developer's MBP is overheating for inherent reasons, or it's just that his mac is full of junk

borkdude12:07:30

it’s fun that a 13" MBP now has the power (4 core i7) of my current 15" MBP (2015) at the weight of a Macbook Air.

henrik12:07:12

@vemv Dave Lee, who alerted everyone to the initial problem, released an follow-up video where he tests the machine after the patch: https://www.youtube.com/watch?v=UTguywiC9aw

borkdude12:07:33

It’s good to see there are improvements over the last batch. Now if my current MBP breaks, I’m not forced to buy a mediocre device.

valtteri14:07:34

IME memory is much more important than high cpu clock speeds in ‘normal’ dev work. You rarely have all CPU’s grinding at 100% but you often have all kinds of programs open eating up the memory.

valtteri14:07:54

In case of MBP, I’d spend the +500$ to double the memory from 16->32 gigs rather than i7->i9.

lilactown14:07:39

i'm still salty that I bought the 16gb version 6 mo ago... I knew I should have held out for the 32gb version >_<

eggsyntax14:07:45

re "You rarely have all CPU’s grinding at 100%" Depends what you're doing 😉 It's not all that uncommon for me to have most CPUs pegged at 100% for a while. That said, ++mem as well. I moved to 32GB when I switched to a Linux box last year (Thinkpad P51), and for once I almost never max out my mem use (although I do go over 16GB routinely).

bherrmann14:07:40

I know the intellij user population doesn't overlap with clojure users entirely... but still... clojure/clojurescript sure is a small part of the pie! This is from https://www.jetbrains.com/research/devecosystem-2018/ I guess we are solidly 1% ers.....

vemv15:07:09

@valtteri my question was focused on number of cores, not clock speed. Core count should decrease resource contention (each process doesn't context-switch as much), and mitigate the effects of demanding processes (e.g. a process is stuck in a loop)

3Jane15:07:27

@bherrmann I wonder how this looks like when you compare in specific niches

3Jane15:07:14

Eg. a person who does web might be doing PHP/frontend-Javascript, or Python/frontend-Javascript, or node/frontend-Javascript

3Jane15:07:13

if we count all of them as “does javascript”, of course it’s going to be high on the list of use, since there’s no other real option

3Jane15:07:49

likewise shell scripting languages (though I guess these have some competition from Perl and maybe Python), and sql.

valtteri15:07:49

@vemv yep, my comment wasn’t directed to anyone. I just felt urgent need to shout that out loud for no rational reason. 🙂 I also think more cores is a. 👍

🙌 4
henrik15:07:49

Too relatable for comfort.

henrik15:07:12

Replace rakes with a mine field, and you have me trying to use Vim or Emacs.

manutter5119:07:22

Cool, I just got a blackmail threat in my email. Sounds like it’s going around. https://www.reddit.com/r/Scams/comments/8gsjba/the_blackmail_email_scam/

bherrmann19:07:03

Got an email from my aunt... on vacation passport/wallet stolen.... can someone please help.... email address was <mailto:[email protected]|[email protected]> - it looked legit.... naturally nothing about it was legit....

manutter5119:07:00

Return address on mine was too :thinking_face: