SD Card reader speed

[skanter]

My Dell XPS 8940 hs a card reader, but slow at 25 MB/sec. So I bought this reader:


uni SD Card Reader, High-Speed USB 3.0 to Micro SD Card Adapter, Aluminum Computer Memory Card Reader Dual Slots, for SD/SDXC/SDHC/MMC/Micro SDXC/TF/Micro SDHC | Laptop, PC, and More https://a.co/d/dTVLFfW

My Dell XPS 8940 has 3 USB 3.1 gen 1 ports on front. I am using a Sandisk extreme 200MB card, UHS-1, but only getting 60-70 MB/sec transfer speed. Shouldn't it be faster? Where is slowdown? Thanks in advance…

 

[skanter]

Your a6400 can't fully utilize UHS-II card.

I use UHS-II V90 SD, and the practical transfer rate to a PC is about 120-180MB/s using a dual CFExpress A & UHS-II SD reader.

Yeah, I just asked perplexity - a6400 cannot make use of UHS-II, only UHS-I speeds. So not worth of it for me to get UHS-II card and reader.

No problem, transfer speed is fine at 50-90 MB/sec.

That’s a good number of images/sec.

Note that there’s sometimes confusion between megabytes and megabits when comparing transfer speeds. (MB = megabytes, Mb = megabits).

Thanks for the correction.

If I recall, you gave the correct units.

I’ve hardly ever measured actual transfer speeds on my systems, since everything is “fast enough”, including SD cards on a Dell XPS 8940 (now retired*).

My system has a readout when transferring files, showing speed in real time. With internal SD card slot, it was stable at around 25 Mb/sec. With new reader, it varies continually from 50-90 Mb while transferring. I wonder why?

My XP 8940 is running Lightroom beautifully since I added a used GPU (RTX 3060ti) recently as it had been using the internal GPU. I added some SD HDs wen I bought it 5 years ago.

I do have a laptop with a microSD slot that is occupied with a 32GB card just for completeness. - It would only be of any use if I wanted to backup some local data while travelling.

* Power supply failure. Computer no longer needed so kept for spare parts.

My XP 8940 has a 500W PS, some have only 300W. I hope I can replace it if it fails.

500W in mine too, and it’s been a good performer until the PS died. I decided to replace it with a laptop and keep the XPS for spares. I have another Dell of a similar model.

Rather than spend good money on the Dell, I’ll be getting something new when it’s needed, probably an i5 which seems to be the sweet spot at the moment.

I’ve used Dell Studio XPs for many years, replacing some failed parts like PS. My last one lasted 10 years. I’m under the impression that the 8940 PS is not easy to replace.

Are you referring to buying a replacement, or physically doing it?

If it's the former, and you're willing to use eBay, there are many 8940 PSUs for sale. Amazon has some too, but their search returns quite a few that look like a different form factor.

I would physically replace as I’ve done in other systems, but right now it’s fine.

 

[skanter]

500W in my 8940 too, and it’s been a good performer until the PS died. I decided to replace it with a laptop and keep the XPS for spares. I have another Dell of a similar model.

skanter wrote:

I’ve used Dell Studio XPs for many years, replacing some failed parts like PS. My last one lasted 10 years. I’m under the impression that the 8940 PS is not easy to replace.

Are you referring to buying a replacement, or physically doing it?

If it's the former, and you're willing to use eBay, there are many 8940 PSUs for sale. Amazon has some too, but their search returns quite a few that look like a different form factor.

I wasn’t impressed by my search for replacement OS units for the 8940.

Did you mean PS units?

Physical replacement would not be a problem, I have had the PS out to check for obvious faults. It involved a couple of fixing screws and the usual connections.

If it’s the right form factor it’s easy.

 

[number_5]

The transfer rate is at the mercy of the weakest component.

Yes, exactly. And in my experience the weakest component is often the cable, because without a cable tester there's no easy way to tell exactly what kind of cable you've got.

Most egregious, IMHO, is the fact that you can't assume a USB-C cable is fast. USB-C is a terrific advancement in capability, but it doesn't require cables to support USB 3 connections and it's my impression that your general run-of-the-mill USB-C cables run at USB 2 speeds, a crying shame methinks.

But the cheap reader I just bought has its own cable attached to the reader.

I have some with cable attached and some without. The fact that you can't remove the cable guarantees it' not a charging cable, which might run on USB 2 (480Mbs) speed. After I got home from the trip, I labeled all the data cables.

uni SD Card Reader, High-Speed USB 3.0 to Micro SD Card Adapter, Aluminum Computer Memory Card Reader Dual Slots, for SD/SDXC/SDHC/MMC/Micro SDXC/TF/Micro SDHC | Laptop, PC, and More https://a.co/d/9z3muNp

My SD card reader (SD only) has built-in undetachable cable. It has stated 500MB/s max read, and IIRC, the transfer is still in mostly 120-180 MB/s range with a V90.

 

[number_5]

a reliable 70/min still means 15 seconds per gigabyte. On a twice a year basis, I might come back with 200gb in data across CFE, microsd, and sdhc. It can be slow, but it runs in the background.

Not sure what you mean.

I regularly have cards with over 1000 26MB files that i will process in LR. With the internal SD reader I went away and did something else while transferring, 15or 20 minutes. Hoping to get started sooner with PP.

26G at 15/sec/G = 6.5 minutes. Or 1/10th the time you've spent on this thread, without bothering to google the different interface speeds.

What’s your point (if there is one)?

I think the point was he can google ... sorry, that one was just too easy

 

[ggeinec]

. . . sniooed . . .

I’ve used Dell Studio XPs for many years, replacing some failed parts like PS. My last one lasted 10 years. I’m under the impression that the 8940 PS is not easy to replace.

I'm still using a Dell Studio XPS 8100 I bought in 2010 as I transition to a new PC. Along the way I replaced the PSU once, and HDDs a couple of times out of caution because of age or capacity, not because of problems. Otherwise mine has been rock solid reliable. When I finally retire it I'll miss it the same way I miss an old car that has served me well for many years . . .

 

[skanter]

a reliable 70/min still means 15 seconds per gigabyte. On a twice a year basis, I might come back with 200gb in data across CFE, microsd, and sdhc. It can be slow, but it runs in the background.

Not sure what you mean.

I regularly have cards with over 1000 26MB files that i will process in LR. With the internal SD reader I went away and did something else while transferring, 15or 20 minutes. Hoping to get started sooner with PP.

26G at 15/sec/G = 6.5 minutes. Or 1/10th the time you've spent on this thread, without bothering to google the different interface speeds.

What’s your point (if there is one)?

I think the point was he can google ... sorry, that one was just too easy

Do people still use Google? Try Perplexity, far superior.

--
Sam K., NYC
“I’m halfway between tightrope walker and pickpocket.” HCB
—
__
Smugmug Galleries:
http://skanter.smugmug.com
—
Instagram:
https://www.instagram.com/sam.kanter/

 

[skanter]

. . . sniooed . . .

I’ve used Dell Studio XPs for many years, replacing some failed parts like PS. My last one lasted 10 years. I’m under the impression that the 8940 PS is not easy to replace.

I'm still using a Dell Studio XPS 8100 I bought in 2010 as I transition to a new PC. Along the way I replaced the PSU once, and HDDs a couple of times out of caution because of age or capacity, not because of problems. Otherwise mine has been rock solid reliable. When I finally retire it I'll miss it the same way I miss an old car that has served me well for many years . . .

Dell Studio XPs are great, about 1/2 the price of equivalent Macs with much more flexibility

My last one lasted 10 years, mine now is 5-years and runs fast, silent and beautifully..

 

[skanter]

The transfer rate is at the mercy of the weakest component.

Yes, exactly. And in my experience the weakest component is often the cable, because without a cable tester there's no easy way to tell exactly what kind of cable you've got.

Most egregious, IMHO, is the fact that you can't assume a USB-C cable is fast. USB-C is a terrific advancement in capability, but it doesn't require cables to support USB 3 connections and it's my impression that your general run-of-the-mill USB-C cables run at USB 2 speeds, a crying shame methinks.

But the cheap reader I just bought has its own cable attached to the reader.

I have some with cable attached and some without. The fact that you can't remove the cable guarantees it' not a charging cable, which might run on USB 2 (480Mbs) speed. After I got home from the trip, I labeled all the data cables.

uni SD Card Reader, High-Speed USB 3.0 to Micro SD Card Adapter, Aluminum Computer Memory Card Reader Dual Slots, for SD/SDXC/SDHC/MMC/Micro SDXC/TF/Micro SDHC | Laptop, PC, and More https://a.co/d/9z3muNp

My SD card reader (SD only) has built-in undetachable cable. It has stated 500MB/s max read, and IIRC, the transfer is still in mostly 120-180 MB/s range with a V90.

I was thinking of buying that one with a new UHS-2 SD card, but since my camera could not use the speed I decided to stick with UHS-1 speeds. Fast enough.

 

[Sean Nelson]

The transfer rate is at the mercy of the weakest component.

Yes, exactly. And in my experience the weakest component is often the cable, because without a cable tester there's no easy way to tell exactly what kind of cable you've got.

But the cheap reader I just bought has its own cable attached to the reader.

My comment was more of a general one directed at anyone who's reading the thread.

 

[kelpdiver]

But I'm accustomed to seeing major speed variations while copying single large files from an NVME SSD to a SATA HDD.

Even for a large single sequential write, the HDD is burdened by its design, particularly if fragemented. It could be as simple as the drive cache gets filled (showing spike in write speed), then slower again as it needs to actually put it to disk. Small copies could stay entirely in cache, but not large file, or large file count.

And then there is the question of if Windows File Explorer is even giving us valid data on the copy speed. I wouldn't trust it over a stopwatch and simple division. Let's not make the mistake that displayed data precision actually translates to accuracy.