What's new

Just upgraded from N to AC, reality check (Mac-to-PC backup)...

  • SNBForums Code of Conduct

    SNBForums is a community for everyone, no matter what their level of experience.

    Please be tolerant and patient of others, especially newcomers. We are all here to share and learn!

    The rules are simple: Be patient, be nice, be helpful or be gone!

turnstyle

Regular Contributor
Hi all, I recently replaced my N-class router with an AC.

Overall, Web browsing does seem peppier. However my main metric was a bit disappointing -- perhaps my expectations (and understandings of my bottlenecks) were wrong, I hope you don't mind an explanation...

I have a PC wired by Ethernet to the router, and that PC has an external drive attached via UBS2.

I have a Mac that connects over wifi, and runs ChronoSync to backup the Mac to the USB drive on the PC. The Mac is connected to the PC via a smb: path.

My biggest file that gets backed up regularly is a ~1GB Lightroom catalog file.

Over my previous N router, ChronoSync would generally report roughly a 5-8MB/s transfer.

Over the new AC router, ChronoSync seems to report a 8-9.5MB/s transfer.

My understanding is USB2 is 480Mbit/s -- so 60MB/s.

Is there a way to understand what the bottleneck is? Is 8-9.5MB/s reasonable for backing up from a Mac over AC and smb to a USB2 drive on PC?

Thanks for any reality check!
 
The limit isn't the USB bus, it's the wireless connection. Unless the Mac has an AC radio, you shouldn't expect a huge jump in wireless throughput.
 
The limit isn't the USB bus, it's the wireless connection. Unless the Mac has an AC radio, you shouldn't expect a huge jump in wireless throughput.

Thanks -- the Mac DOES have an AC radio -- it's a "late 2013" 15" retina MacBook Pro -- so it does have AC (and, from what I understand, 3x3).
 
Thanks -- the Mac DOES have an AC radio -- it's a "late 2013" 15" retina MacBook Pro -- so it does have AC (and, from what I understand, 3x3).
So what router are you using?
 
Is the drive externally powered or USB?

Max for USB 2.0 is 35MB iirc. USB-powered probably means even slower?
 
New router is TP-Link Archer C8.

The USB drives are wall-powered, 3.5" -- I believe 7200prm.

So USB2 shouldn't be the bottleneck here?

What about using smb to connect a Mac to a PC? Is there some sort of big conversion hit?

What should I generally expect for over-AC-to-USB2 bandwidth?

Many Thanks!
 
What kind of performance do you get over ethernet on the same application - might be ChronoSync is going along as fast as it can...
 
Now I'm thinking ChronoSync may be reporting oddly (or perhaps operating slower than it should/could be).

I just dragged a 3.15GB file from the PC to my Mac via Finder, and that too 90 seconds -- so that comes out to about 35MB/s -- which is freakishly close to Nullity's post above.

ChronoSync was writing TO the USB drive, whereas the 3.15GB Finder copy was a read -- so perhaps I should try writing the 3.5GB file back TO the drive (in Finder rather than ChronoSync).

Anyhow, I now think I'm set -- sorry for what is probably a false alarm -- and blast that USB2 port! :)
 
Ok, so...

So copying the 3.15GB file from the USB2 drive on the PC to the Mac over AC took 90 seconds -- but copying the 3.15 GB file from the Mac to the same PC drive took 165 seconds, so almost twice as long. Or about 35MB/s to copy from the PC, and about 19MB/s to write to it.

Is that about what one would expect? Thanks again!

The PC is an old Thinkpad from just before they stated adding USB3 -- I thought about getting an ExpressCard USB3 -- but I kind of worry that it might be more trouble than it's really worth?
 
Ok, so...

So copying the 3.15GB file from the USB2 drive on the PC to the Mac over AC took 90 seconds -- but copying the 3.15 GB file from the Mac to the same PC drive took 165 seconds, so almost twice as long. Or about 35MB/s to copy from the PC, and about 19MB/s to write to it.

Is that about what one would expect? Thanks again!

The PC is an old Thinkpad from just before they stated adding USB3 -- I thought about getting an ExpressCard USB3 -- but I kind of worry that it might be more trouble than it's really worth?

I got one of those ExpressCard USB 3.0 cards, increased speed over 2.0, but not by much. Firmware is hard to find but it kinda works with Linux, which is better than expected. I am mostly disappointed that I cannot boot to the ExpressCard...


Your drive, if old or cheap, is working fine, maybe.
 
I got one of those ExpressCard USB 3.0 cards, increased speed over 2.0, but not by much. Firmware is hard to find but it kinda works with Linux, which is better than expected. I am mostly disappointed that I cannot boot to the ExpressCard...

Your drive, if old or cheap, is working fine, maybe.

Thanks for confirming -- on a previous laptop I bought an eSATA card -- when it worked, it was definitely an improvement -- but it just wan't consistent, drives would disappear, and I finally gave up -- so now I'm weary.

Too bad there isn't a USB3 ExpressCard that "just works".
 
Should have checked the review first. Write and read performance are significantly different and in the range of what you are finding.

USB powered vs. wall power makes no difference.
 
Should have checked the review first. Write and read performance are significantly different and in the range of what you are finding.

USB powered vs. wall power makes no difference.


Hey Tim, just to clarify: I'm not using a USB drive attached to the Archer C8 -- the USB drive is attached to a PC which is wired by Ethernet to the router -- was your comment above written thinking I had the USB attached to the C8?

I DID read MANY of the very helpful reviews and articles on this site before I settled on the C8 -- so now I'm a bit confused by your comment here?

(ps: when Nullity asked about USB power, I assumed that's because USB-powered drives run at slower RPM.)
 
Sorry. I thought at some point you were referring to transfer rate to a drive attached to the C8.

As OPs have suggested, run your same transfers using Ethernet to determine the maximum transfer rates.
 
Sorry. I thought at some point you were referring to transfer rate to a drive attached to the C8.

As OPs have suggested, run your same transfers using Ethernet to determine the maximum transfer rates.

phew! fwiw, as it stands now, I'm inclined to believe the problem is that ChronoSync is reporting the speed incorrectly (or perhaps ChronoSync is running slow) -- because a copy through the Finder seems more in-line with expectations.

Here's what I'm seeing with a plain Finder copy:

1) Copy from USB2 drive, attached to Win7 PC, over AC, using smb:, to the Mac -- that appears to be about 35MB/s

2) Copy to the USB2 drive, attached to Win7 PC, over AC, using smb:, from the Mac -- that appears to be about 19MB/s

Just to confirm, do these numbers seem "ballpark reasonable?
 
Hey, a question...

I just set up a new share on the internal drive of the PC (an SSD) rather than the external USB2.

In this case, copying from the PC (so now using an internal SSD rather than external USB2) -- over AC via smb: connection in Finder -- now it took about 60 seconds, as compared to 90 seconds when copying from a USB2 drive.

So that means from 35MB/s to about 52MB/s.

Likewise, writing to the SSD share went from about 19MB/s to about 31MB/s.

Does this mean 52MB/s read and 31MB/s write is roughly as fast as I can expect over AC1750 with the USB2 bottleneck eliminated?

Or should I be faster than that? And could the Mac/Windows difference be responsible for the slower-than-expected transfer?
 

Latest threads

Sign Up For SNBForums Daily Digest

Get an update of what's new every day delivered to your mailbox. Sign up here!
Top