rMBP 256 vs 512 in *speed performance*

[Falundir]

I have been doing a fair amount of research, and haven't found anyone who has done a comparison between the 256 and the 512 in speed. If anyone has seen an article I might have missed would you be so kind as to post it up.

My personal determining factor for the model I choose is SSD speed.

Thanks in advance!

 

[varsis]

I have been doing a fair amount of research, and haven't found anyone who has done a comparison between the 256 and the 512 in speed. If anyone has seen an article I might have missed would you be so kind as to post it up.

My personal determining factor for the model I choose is SSD speed.

Thanks in advance!

Should be the same speed. Both use the 830 samsung controller.

 

[BB.King]

512GB likely a bit faster as it have higher density. Why would SSD speed be a major factor in your decision? You are unlikely to find much faster SSD unless they are in RAID 0 setup.

 

[Freyqq]

512GB likely a bit faster as it have higher density. Why would SSD speed be a major factor in your decision? You are unlikely to find much faster SSD unless they are in RAID 0 setup.

Nope. I'm pretty sure there are just more chips. So, same speed.

 

[Mr MM]

Nope. I'm pretty sure there are just more chips. So, same speed.

not necessarily, larger drives have higher speeds, 256gb and 512gb usually dont have that much difference in speed though, not that you were going to notice

 

[Freyqq]

not necessarily, larger drives have higher speeds, 256gb and 512gb usually dont have that much difference in speed though, not that you were going to notice

depends on the drive. All i'm saying is that usually 256 and 512 only have a difference in # of similar chips

 

[ljx718]

512GB likely a bit faster as it have higher density. Why would SSD speed be a major factor in your decision? You are unlikely to find much faster SSD unless they are in RAID 0 setup.

youre thinking of old plater drives. SSDs just differ in number of chips. if it was different speeds but the controller and chips were the same, it would be the manufacturer just artificially limiting the speed of the lower capacity one

 

[Fry-man22]

youre thinking of old plater drives. SSDs just differ in number of chips. if it was different speeds but the controller and chips were the same, it would be the manufacturer just artificially limiting the speed of the lower capacity one

Some lower capacity SSDs (64/128GB) do not have fully populated memory banks on the controller. Since this is where the transfer to/from the memory occurs, drives that do not have fully populated "banks" will be slower than the larger ones where they are fully populated.

I have never read a review where the 256GB example of a product did not have fully populated banks.

 

[melterx12]

Nope. I'm pretty sure there are just more chips. So, same speed.

Yes, more chips, however this is precisely why a larger capacity drive is faster.

On it's own, a single chip inside an SSD is actually pretty slow. The way SSD manufacturers achieve high speeds is by putting multiple chips on a PCB and using them in RAID0 with the internal SSD controller.

A higher capacity drive likely has more chips than a lower capacity drive, and more chips in RAID0 = more speed.

Only in the very high capacities does an SSD manufacturer use higher density chips, because there simply isn't enough space to pack more chips on the PCB at that point

 

[Falundir]

Thanks kindly for the replies all.

To the question of why it matters to me, I do computer forensics, and a 2% increase in speed can mean hours if not days of less processing time.

The reality is evidence is processed using external RAID storage, but I like to have the temp storage on a local SSD so you are utilizing different sources for each task. Simple concept really. Evidence on RAID, Temp on SSD, Results on single drive in a 'toaster' aka drive dock.

 

[Slivortal]

Thanks kindly for the replies all.

To the question of why it matters to me, I do computer forensics, and a 2% increase in speed can mean hours if not days of less processing time.

The reality is evidence is processed using external RAID storage, but I like to have the temp storage on a local SSD so you are utilizing different sources for each task. Simple concept really. Evidence on RAID, Temp on SSD, Results on single drive in a 'toaster' aka drive dock.

If you're THAT concerned about performance, shouldn't you go for the 2.7 GHz model for the 8MB L3 cache anyways (which will actually provide real gains as opposed to the 2%)? And since that processor requires the 512GB model, isn't your problem solved?

Larger SSD's is one of those things like 2.6 vs 2.3 GHz. Neither will almost ever do anything, but if you're in a position where one matters, the other probably will as well.

 

[DiGriz]

Up to a point larger drives use more chips (which can be accessed in parallel), and are therefore faster. This effect stops at about the 256GB point for this generation of SSDs. However, if Apple are using the Samsung 830, then the 512GB drive is on average (according the Anadtech's benchmarks) fractionally slower than the 256GB version. Not by any noticeable amount, but a small amount nonetheless.

 

[pandamonia]

Actually 512GB are sometimes slower than the 256 versions.

 

[leman]

Thanks kindly for the replies all.

To the question of why it matters to me, I do computer forensics, and a 2% increase in speed can mean hours if not days of less processing time.

The reality is evidence is processed using external RAID storage, but I like to have the temp storage on a local SSD so you are utilizing different sources for each task. Simple concept really. Evidence on RAID, Temp on SSD, Results on single drive in a 'toaster' aka drive dock.

Quite honestly, I don't think that its such a good idea to use your laptop for computations which take several weeks to complete. This is something which should be done on a stationary well-ventilated server.

 

[Puevlo]

Thanks kindly for the replies all.

To the question of why it matters to me, I do computer forensics, and a 2% increase in speed can mean hours if not days of less processing time.

The reality is evidence is processed using external RAID storage, but I like to have the temp storage on a local SSD so you are utilizing different sources for each task. Simple concept really. Evidence on RAID, Temp on SSD, Results on single drive in a 'toaster' aka drive dock.

Just start the test a few hours earlier. Then you'll be finished on time.