• 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Forums > Back > Samyang 12mm F2 NCS CS - asymmetric coma?
#21
If you see the processor load "far less than 100%", it's not multithread. The apps I have and know for sure are multithread, show it by going to 100% on each of the 4 possible channels. Other apps I know for sure the are not supporting more than a single processor are usually managed by Mac OS and you see in total not more than 100 % for all channels.
  Reply
#22
I know, I was referring to a tool which gives the number for each core. iStats, instead, typically gives numbers around 500%/600%, 800% being the max (4 cores with 2 threads each one), thus with peaks at 75% of the total power. I/O speeds aren't over 15/20MB/sec, while the SSD is capable to do 1.5GB/sec and better.

stoppingdown.net

 

Sony a6300, Sony a6000, Sony NEX-6, Sony E 10-18mm F4 OSS, Sony Zeiss Vario-Tessar T* E 16-70mm F4 ZA OSS, Sony FE 70-200mm F4 G OSS, Sigma 150-600mm Æ’/5-6.3 DG OS HSM Contemporary, Samyang 12mm Æ’/2, Sigma 30mm F2.8 DN | A, Meyer Gorlitz Trioplan 100mm Æ’/2.8, Samyang 8mm Æ’/3.5 fish-eye II | Zenit Helios 44-2 58mm Æ’/2 
Plus some legacy Nikkor lenses.
  Reply
#23
you might be getting confused between Mega bits and Mega bytes - 1M byte = 8M bits (8 bits in a byte).

 

So 1.5GBits/sec is equal to 23.5 M byte / sec.

 

Also the manufacturers tend to specify the data rate supported by the interface, however there will be overheads involved such as packet id, checksums for example, so the actual transfer rate of useful data is lower.

 

Of course I could be wrong and its pants... Smile

  Reply
#24
Also, other hardware could be used for processing, Aperture used the graphics card extensively. I'm unsure what for though.

  Reply
#25
23.5Mbyte/sec would be quite poor for an SSD...  Smile (even for an HDD) . All my figures were in bytes/sec. The PCIe SSD in the Retina MacBook Pros really reads at 1.6Gbytes/sec and writes at 1.0Gbytes/sec (confirmed by a quick-n-dirty test). In other words, they can read the dump of a DVD in a matter of seconds.

 

Also for me is partially obscure what use LR makes of the graphic card (it does, also because there's an option to deactivate it). For sure there's a visible effect when you touch the editing sliders. I'm unsure on whether it also uses it for operations such as batch processing.

 

PS Measured on an encrypted partition. On the newest MBP it's unclear to me where the bottleneck is, but since the CPU has hardware acceleration for AES encryption, and benchmarks in RAM show a speed of 4/5 GBytes per second, I presume encryption - at least the one provided by HFS+ - doesn't impact speed.

Code:
[Mistral:~] fritz% dd if=/dev/zero of=test bs=1024k count=10240
10240+0 records in
10240+0 records out
10737418240 bytes transferred in 11.366524 secs (944652759 bytes/sec)
[Mistral:~] fritz% sudo purge (clears RAM cache)
Password:
[Mistral:~] fritz% dd of=/dev/null if=test bs=1024k count=10240
10240+0 records in
10240+0 records out
10737418240 bytes transferred in 5.854027 secs (1834193484 bytes/sec)

stoppingdown.net

 

Sony a6300, Sony a6000, Sony NEX-6, Sony E 10-18mm F4 OSS, Sony Zeiss Vario-Tessar T* E 16-70mm F4 ZA OSS, Sony FE 70-200mm F4 G OSS, Sigma 150-600mm Æ’/5-6.3 DG OS HSM Contemporary, Samyang 12mm Æ’/2, Sigma 30mm F2.8 DN | A, Meyer Gorlitz Trioplan 100mm Æ’/2.8, Samyang 8mm Æ’/3.5 fish-eye II | Zenit Helios 44-2 58mm Æ’/2 
Plus some legacy Nikkor lenses.
  Reply


Forum Jump:


Users browsing this thread: 5 Guest(s)