Discussion WTF Samsung Magician

[LOUISSSSS]

How do they figure???

 

[extide]

Seems about right to me. Toshiba drive looking a bit slow in sequential for an NVMe drive but I believe that one is only PCIe x2 -- so those speeds are appropriate.

 

[Billy Tallis]

Only thing that looks wrong to me is that the NVMe drive performance is graphed on a different scale from the SATA drives. The raw numbers seem reasonable.

 

[LOUISSSSS]

right... why is the scaling all off and skewed for the Samsung drive i have? (an old ass 840 from 2013)

 

[VirtualLarry]

Scaling is clearly as per the maximum theoretical sequential bandwidth for the interface. So the SATA (Samsung) drives are achieving 90%+ theoretical SATA bandwidth, but the NVMe is only around maybe 30-40% max theoretical bandwidth for NVMe drives.

Makes sense to me.

 

[aigomorla]

Its because magican is using your system ram as a buffer, and artificially tweaking your results.
You can clearly see that the numbers being shown is your RAM's speed.

 

[EXCellR8]

graphs!!!

 

[extide]

Its because magican is using your system ram as a buffer, and artificially tweaking your results.
You can clearly see that the numbers being shown is your RAM's speed.

No, it is not.

 

[aigomorla]

Welp i read his gif wrong, so Larry is correct.

Rapid does artificially tweak the benchmarks, so i misread his NVMe as being the SATA.

Thread here....

Enabling RAPID Mode in Samsung Magician?

Running Samsung Magician's own Performance Benchmark on a 850 EVO, here's how it affected my scores (all with 'OS Optimization' set for 'Maximum Reliability'): Without RAPID Mode: Sequential Read: 549 Sequential Write: 528 Random Read: 96839 Random Write: 88940 With RAPID Mode...

forums.anandtech.com