Any idea what processor Apple is using in the MacBook Air? It sorta matches the 8200Y, but it has UHD 617 graphics and a 1.6GHz base speed. Custom processor?
-
-
-
A tweaked speed bin. We're waiting on an email with full confirmation
End of conversation
New conversation -
-
-
What's a GB4 in this case? Didn't see that part of the event.
-
Geekbench 4. Has an iOS and Windows version of the benchmark. Used in a lot of cross platform comparisons
-
Ohh, for some reason I didn't make the connection that GB = Geekbench
- 1 more reply
New conversation -
-
-
I'd like to see you do a write up on this, have it as an article.
-
+1. However, I think a good test would be to do something like: I want to do "x" (turn a photo 180 deg, search a multi-gigabyte string array [sorted & unsorted] for a value, perform complex aggregation on a multi-GB 2D array [basically approximating a database table without /2
-
1\ introducing 3rd party database implementations into the mix, or maybe just search a B-tree]) so let's create some C/C++ programs to do it, and then compare the programs (using different compilers, of course, because they each have their own little niggles) PLUS our /3
-
2\ hand-tuned assembly programs to do a proper comparison. Dude, that's a metric ton of work if one wants to do it properly, because one needs in-depth knowledge of each hardware platform (Intel, AMD, and ARM) & operating system/software ecosystem PLUS years of experience /4
-
3\ doing it. <<<EOF>>>
-
1/\I'm still against cross comparing chips w/ different archs because they'll never run the same binary. How else would you go about creating the same exact conditions on ARM & INTEL chips that will put the different-archs argument to rest? HOW?!
-
2/\ let's just stick to comparing Intel with AMD, NVIDIA with Radeon, and ARM chips with other ARM chips... Because if we start cross comparing, we're gonna lie to ourselves about the true performance of the processors...
-
Nope. Not if one sets a uarch-independant goal of a task that needs to be accomplished, like I said in my tweets, and then codes the same algorithm to do it.
- 8 more replies
New conversation -
-
-
using GB to compare anything is pure idiocy. It is an inconsisten benchmark that is designed to be "pay to win".
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
SPECcpu2017
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
You mean GB in particular? or do you include SPEC, etc?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
The best explanation of this is a 4 year old post on RWT.... Surely you could throw something together that is better and more up to date with its critique!
-
Yup, that would be great :)
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.