• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

WinRAR 4.20 gets multi-threaded support

Page 3 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Status
Not open for further replies.
woops, Nehalem.
Looks like we can meet the Nehalems. About time.

🙂 970 is gulftown 920 nehalem

IPC, and memory bandwidth would be equal just different core count and more L2 Cache and AES!

And not surprised by BD's results when there are apps that can acutally use its cores performance is good, its just the single threaded performance where its lacking.
 
Last edited:
Yeah I'm sure. 6 Cores vs. 4 Cores. :biggrin:
To your advantage, you were running a much higher overclock, let alone the incredible efficiency of Ivy Bridge architecture. Maybe, I had better memory timings (1600 @ 7-7-7-24). Re-run with those if you can and see if it makes any difference.
 
Last edited:
Upon further inspection, it looks like 7-zip is smart enough to figure out that some files in one folder were duplicated in another, so it somehow linked them to the same data area. Wow 😀

WinRAR will likely do the same thing if you choose the 'solid archive' option

solid archiving combines all the files together for compression purposes

pro: it can take advantage of duplication between files
cons: if one part is corrupt, you can lose all the files in the archive instead of just those affected by the corruption (it's more fragile)

of course that is where WinRAR just trounces 7Zip, by being able to include a recovery record
 
WinRAR will likely do the same thing if you choose the 'solid archive' option

solid archiving combines all the files together for compression purposes

pro: it can take advantage of duplication between files
cons: if one part is corrupt, you can lose all the files in the archive instead of just those affected by the corruption (it's more fragile)

of course that is where WinRAR just trounces 7Zip, by being able to include a recovery record

I use Winrar with solid archiving but I just tell it to add a 5% recovery record.

If any file in a solid archive is damaged, it will be impossible to extract all files which follow the damaged area. Thus if a solid archive is stored to a potentially unreliable media, it is recommended to make use of the recovery record.
 
i3-530 – 2.93 …

HT off … 1684
HT on … 2880 … … +71%

Nice boost from hyperthreading on dual cores.
 
Core i7 3770k (IB) - 8,236 kb/s (8000-8200 kb/s over 3 tests)
Core i7 2630QM (SB) - 5,527 kb/s

Both at stock. I tested the 3770k just to make sure, as it seemed a bit high compared to the rest of you guys, but I would not falsify my results. Judging by the multithread use, I'm going to assume that the 3770k ran at 3.5ghz, and the 2630QM ran at 2.0ghz.

*The laptop was plugged into the wall.
 
It has been multi-threaded for a while but not as efficient.
With 4.2 I'm seeing about 55-65% utilization (24 threads).

rar42.gif


In the past disabling HT on hex core systems yielded higher throughput.
 
It has been multi-threaded for a while but not as efficient.
With 4.2 I'm seeing about 55-65% utilization (24 threads).

rar42.gif


In the past disabling HT on hex core systems yielded higher throughput.

What processor in that system ruby ?

24 threads sounds like a server dual socket board perhaps?
 
Xeon X5690 at 4.5GHz (two processors)

Ah thanks for the reply I was actually expecting a higher score from that setup. But the X5690 is based on westmere/gulftown and we don't know what your memory speed is and the memory configuration on that server. And based on the testing that AtenRA did this benchmarks loves memory speed and bandwidth.

So it doesn't look like Winrar was efficiently using all 24 cores or maybe the benchmark isn't up to par for that many cores and it may require testing on a large archive instead.

I'm still waiting for someone to post something from SB-E.

But interesting results so far.
 
Memory speed is 1600MHz with CAS 7 timings, 1T.
It can go faster (memory) and is reflected in the score. This system was around 6500 on the previous Win RAR (4.01) that was installed.
 
It could not really make use of many threads (3+) until the recent update.

We still haven't seen any data on 2011-based systems. Anyone?

Hmm.. I tried the built-in benchmark of version 4.01 and it loads 70%+ of my 2500K. I am also pretty sure that if you disable cores from the BIOS you will get lower scores.

Maybe they made it more efficient, but older versions too were able to use many cores.
 
Status
Not open for further replies.
Back
Top