Slashdot Mirror
A Memory Card Torture Test
An anonymous reader writes "Would you buy a Ferrari and put regular gas into it? I don't think so. So why are most of us buying expensive digital cameras and using cheap memory cards? If you want to find out how much better a high speed memory card is, check out this group test of high capacity compact flash and SD cards."
The Rob Galbraith DPI website has a huge database of performance with various cards and various cameras. I use this as a benchmark for deciding when I need a new CF card vs. the Camera I have, and the family of camera I'd love to upgrade too one day.
As a rock-in-roll Physicist once said, No matter where you go, there you are.
What's the difference between regular gas and this special stuff? Does that mean when you buy a Ferrari you spend half you life looking for Ferrari-approved filling stations?
(These are serious questions ...)
Rich.
libguestfs - tools for accessing and modifying virtual machine disk images
Hint: skip to page 18 for the conclusions.
You don't get any more professional than padding your 3 page article to 19 pages with lines like this (from the conclusion):
Yeah, you could say that. One of those things was my patience.
Stupid sexy Flanders.
When I bought my camera 6 months ago, I searched and searched, and found there was simply no way to know what performance to expect from a given card / camera combination. Labelings like "32x" apparently don't mean a whole lot, the same card doesn't work equally well in all cameras, packaging and labeling are not changed when the card is re-engineered, and there are so many different cards available that no benchmark table is even nearly complete - often there's no overlap at all between the cards used in a benchmark and the cards available from a chosen vendor.
If you take 10,000 photos between taking a "once-in-a-lifetime" photo and backing it up onto a tougher media, you pretty much deserve to lose all your work. The biggest loss of digital camera images are caused by loss/theft of the camera, and user error (accidental deletion). Media failure doesn't even register on the scale.
Athletic Scholarships to universities make as much sense as academic scholarships to sports teams.
More of a high end performance test.
There was one proper torture test done by the UK Digital Camera Shopper magazine where they dipped in cola, run through a washing machine, dunked in coffee, trampled and then for sport hit with a sledgehammer and then nailed to a tree. They didn't survive the last two tests though...
Wonderfully resilient stuff I'd say.
Couldn't find the article freely availabe on the mag, just a ref at BBC news: http://news.bbc.co.uk/2/hi/technology/3939333.stm
"So why are most of us buying expensive digital cameras and using cheap memory cards?"
Because if your camera can write 1MB/s, it doesn't matter if your memory card has theoretical write speed of 1MB/s, 2MB/s, 4MB/s or 10MB/s. You will get 1MB/s of write performance in any case.
It's interesting that they tested memory cards with Canon EOS 1D Mark II camera that costs $3500. I wonder how the results would look if they would've used $350 camera instead.
You'd think cards developed to the same spec would have equal performance. Is that really not the case with SD or others? Interesting article.
I don't know about SD cards, but CF cards are compatible as IDE devices, which itself has various specs with varying performance. Various PIO and DMA modes, etc. This would be like comparing hard drives and then saying, "You'd think drives developed to the same spec would have equal performance".
Some cards are built using high density, low speed, low durability CF, while others go for lower density, high speed, high durability CF and multiples of them in one card. Some newer fast cards employ DMA modes over PIO. Also don't forget, the spec itself is not always the bottleneck, so individual models can vary in performance up to the limit of the particular spec used.
War crimes, torture, lies, illegal spying... Would someone give Bush a blowjob, already, so he can be impeached?
Comment removed based on user account deletion
Some specs allow for variation in performance. For example, if the raw performance of the memory cells is expected to increase significantly, the spec will be designed to allow for the highest speed expected to ever be achieved (or the highest speed economically feasible), but to allow devices to negotiate a slower speed by doing things such as inserting delays. i.e. the spec defines compatibility and not performance, AS LONG as performance can be negotiated to be the lowest common denominator of two devices.
I have heard stories of some of the highest speed cards breaking in older readers, perhaps the autonegotiation was designed with the assumption that cards would always be slower than a reader's capabilities, or those readers aren't fully meeting the specification and no one noticed.
retrorocket.o not found, launch anyway?
You can't put regular gas in a Ferrari? What's the difference between regular gas and this special stuff? Does that mean when you buy a Ferrari you spend half you life looking for Ferrari-approved filling stations?
Someone (with a lot of money) bought one of Schumacher's old F1 cars and yes, it was contractually required that the car only be run on a specific brand of fuel.
The article summary is pretty oblivious though- you run the octane your car requires, 95% of the time. Gasolene companies love to make you think that filling up your low-compression engine (that requires 89) with 94 octane will make it faster, or "clean" it more. All grades of gas from the same brand have the same level of detergents, generally...furthermore, each kind of detergent is good at removing certain deposits but leaves others, so you're actually best off rotating which brand you fill up with. If you're obsessed about it, just pop in a bottle of Techron cleaner one tank before your next oil-change; it's the stuff BMW, Audi, and others recommend, though they'll charge you a lot more for Techron in a BMW or Audi bottle.) Also, most gas is delivered from port by a distributor that slosh-mixes in a bottle of stuff that "makes" the gas Exxon, Shell, Hess, BP, whatever. When a supertanker crosses the ocean, it doesn't have a "Shell" crude compartment and a "Exxon" crude compartment, etc. It's all the same stuff, a commodity...even though Shell likes to run commercials saying their gas meets manufacturer standards blah blah blah. EVERYONE's gas does, because EVERYONE's gas comes from the same damn crude, gets refined at the same places, and distributed by the same companies.
This is similar kind of "inadequacy" based BS. High end digital cameras have large buffers in part because flash memory is so effing slow; a Nikon D70 has enough buffer for something like 40 full resolution JPEG shots! Running a slow memory card in them won't harm them, damage them, etc etc. There are other factors to consider as well- my canon 10D has a 9 shot buffer for RAW shots, and some sort of in-between buffer for writing them to the card. I used to hit the end of the buffer all the time, because I never noticed that it wouldn't process the buffer while the shutter was held half-down in the focus position. Talk about a design flaw- but knowing that, I kept my finger off the shutter button whenever possible if the buffer had anything in it (displayed in the viewfinder) and the problem disappeared.
As someone who has shot with a semi-pro dSLR for more than two years, I can summarize that article in one sentence: "if you need to shoot images as fast as possible and have a camera with a limited buffer, buy the fastest card within reason, only if Rob Galbraith's tests show it'll make a substantial difference. Otherwise, buy a reasonably heard-of brand with a decent warranty in case it stops working." Why? Because just like with the gas, under the label you'll often find exactly the same thing- and only a very small number of people actually NEED the extra speed of a card that costs 50%+ more.
Oh, last piece of advice: don't buy huge memory cards. Three reasons: 1)you pay more per MB, usually. 2)You put all your eggs in one basket- if you drop a card and step on it, accidentally hit "erase all", or loose it... you get the idea. 3)"Photo tanks" with laptop hard drives offer MUCH cheaper $/GB storage. You could shoot 2,3,4,5GB/day in RAWs on a big vacation and still not fill the smallest of these widgets after a week. Buying one without a drive and putting in the old laptop drive you've got hanging around from an upgrade (provided it's not too power-hungry) is the way to go, as even 30-40GB is a BOATLOAD of space for digital photos.
Oh, and should you be on a trip- bring a few DVD-Rs, and burn the files to one or two if you really want to have the photos. Laptops get stolen/dropped/lost/seized/whatevered, and you can be absent minded / mistake-prone about transferring photos after a week of fun in the sun (aka rm -rf * type mistakes). Put one set in your suitcase, another set in your SO's/friend's/etc.
Please help metamoderate.
That's actually determined more by the USB cable you use. Thinner cables are gonna give you better color rendering. Of course if you work in black and white, it doesn't matter. Go ahead and use the cheap stuff. But with color, you don't want multipath blurring your color signals together. And this gets even more important as you shoot multiple frames per second.
And the same thing goes with your storage media. If you work in high resolution color, you need a RAID. That way, you can spread the put the color streams on different physical media to prevent color-bleed. This is even more critical with digital photography, just a one-bit bleed from one pixel to another can ruin a great photo.
So, by all means, get a cheap card if you are going analog black-and-white. But you get what you pay for if you are shooting high-res digital color.
I am not a crackpot.
I currently have two CFcards for my camera, a cheapie that came free with the camera & a SanDisk Ultra II. The SanDisk Ultra II was about twice the price of the cheapie memory, but it'll also write about twice as fast. The Extreme III, however, is what SanDisk are currently pushing as their fastest highest-tech card for your camera, and loads of people buy it. Check the table, however, and you'll see it's only a couple of percent faster in my camera... and at twice the price, of course.
So this is why the Rob Galbraith tables are more useful than some 19-page review full of ads - you can just glance down the page & easily compare the brands that your supplier offers for a real-world comparison and see if they're worth the price.
Stroller.
Another (and I think better) comparison site is here and it has also compares different cameras in conjuction with different cards, which is fantastic if you have one of those cameras. Even if you don't, you can tell whether the card is fast.
Precisely. I was vaguely interested in the article - will a more expensive card improve the shooting speed on my camera? I wondered, or more precisely - would it reduce the delay between being able to take pictures?
Page 2 of the article: "many of our digital cameras have limited write speeds too, so the full potential of these so-called high-speed cards will be restricted.".
So nothing to see here, move along.
My maths may be off, but that means your car gets 25 mpg. When that answer appeared on my calculator, I literally laughed out loud. In the UK we'd refer to that as rubbish efficiency (trends for ludicrous urban SUV usage notwithstanding). 15 years ago I had an ancient piece of crap Morris Minor [wikipedia.org] that did 35+ mpg, ffs.
A UK Gallon & a US Gallon are different.
A UK Gallon is 1.2 * US Gallon.
You are getting 35 MPG(UK) is the same as 29 MPG(US).
Still the Morris minor seems more efficient