Jump to content


Member Since 10 Sep 2008
Offline Last Active Jul 04 2017 04:29 PM

Posts I've Made

In Topic: Atomic Superpi Scores V5

30 June 2017 - 05:21 PM

Well yes. But if you're going to be calculating Pi (or anything) in 2017 for sciencetic stuff are you going to be using x87 or the latest accelerated instruction sets?


You make a good point regarding comparitive results. And of course that's needed. But equally you want benchmarks to reflect real world capability, if you can exploit AVX-512 on one CPU to make it legitimately process something faster you're not showing it's full real world potential by not using it. They may be comparitive, but not the whole story. There's room for both methods imo

In Topic: AtomicMPC 100th Issue Collector's Edition

30 June 2017 - 04:07 PM

That's fair enough.

Yeah space is basically the problem. As a result you just want to make sure all the highlights are touched on and usually by then that's all there's room for.

Sometimes there's just a single page to look at the architecture and platform by itself. I did the X99 stuff when it launched, iirc I got a somwhat generous 3-4 pages to cover just the platform's new features/architecture etc. Did pretty well with it. That's probably the maximum you'll see these days unfortunately.

In Topic: AtomicMPC 100th Issue Collector's Edition

29 June 2017 - 04:16 PM

To answer the OP. I'd like to.

Closest currently published mag in my mind is PCTA but it's more average consumer focused, which is probably why it still exists. But much of it is stuff I'd expect from Atomic in its day.

PCPowerplay is another. I've never read it but I gather it's more gamer oriented. The hardware reviews I've done for them were shorter in length and thus more to the point. Coulnd't go too far in depth.



On the plus side, nothing has actually come up to replace it in the market.


There is no mag that looks into things like architecture, interfaces, bus speeds..... you name it....

If there ever was a hole in the market, and someone wanted to re launch a mag, it's still a hole in the market.


I doubt it'll be filled, but it's there.


I freelance for PC & Tech Authority. I've had the chance to do some in depth articles where we get to go into all that stuff. Usually happens whenever new CPU/GPU lineups launch.

Next issue there'll be in depth articles around Sky/Kabylake-X and X299 motherboards tests etc.


PC centric only coverage of architecture and things I doubt would be enough to make a whole mag worth it. If you're talking any silicon chip in any segment, then maybe.

As it is I do a page each issue with CPU/GPU news where I can touch on those subjects.


The problem with magazines is they are old before they come out, supposedly it takes 2 month to make an issue, by that time it is old news

Yep, not much you can do about time to market. Although it's really only a single month to publish (monthly issues). Even then my few recurring pages usually only take 3 or so days to pump out, with product reviews or round ups adding say an extra week on top of that. With multiple writers doing full time and many freelancers contributing I doubt it really takes the full month to make the content.


Printing and distrubution takes a while. From what I see it's roughly 2 weeks from submission to the printers for it to get into your hands.

In Topic: What type of gamer are you? (Research Study)

29 June 2017 - 02:45 PM

Reclusive Tactician



Immersion Engineer



Are what I got. Spot on really.

What did you get?

In Topic: Atomic Superpi Scores V5

29 June 2017 - 02:11 PM

Nice Sora. Updated your SuperPi and ycruncher scores in OP


Re yCruncher.


I've just been doing some side-by-side BMs with my Haswell i7-4790 (3.6-4.0) vs an Ivy Bridge i7-3770 (3.4-3.9).  Only a generation apart, in fact about 14 months.

With SuperPi they were almost the same, wPrime the earlier one was actually a touch quicker in the 4T 32M one though that was bare desktop vs open browser and other stuff going on.


The surprise packet is yCruncher though.  My Haswell almost twice as fast.  I suspected it might be writing to HDD which could severely handicap a mech vs SSD but it seems it doesn't.

Just goes to show the improvements in the maths processing as time goes on.  I've noticed it before with video encoding, the first gen i7s simply whip the arse of the Core 2 at similar frequencies.

Similar story with some types of work as you get newer i- generations.

Yeah the guy that makes ycruncher makes bespoke binaries for each architecture to make the best use of the intruction sets and hardware accelerated logic available.

Even on the same architechure but with a newer version of ycruncher I've seen large performance gains as he improves his algorithms and tries to make the calculations happen as efficiently as possible with the hardware resources available.


Which is a better way of benching imo, it delivers the real full potential of the chip rather than running the same thing on every CPU the same way, which might not be the best or fastest way of executing it. Like SuperPi and it's long outdated x87 intructions, newer and better methods come along over time. Doing a 1M superpi quivalent run in ycruncher (0, 1, 0, 20), my work (Zen) PC can do the math in 0.1s as opposed to 11.9s.


RE HDD writing, the ycruncher compute time is before it writes the results to disk, so that doesn't affect the score. Only the start-to-end wall time is affected by the mass storage sub system.



Always a good read.

If you see his March 2017 update, most binaries copped a +5% speedup, Zen was added and got a 14% increase too.