Slashdot Mirror
Adobe Says PCs Are Preferred
Father Of Free Choice writes "Abobe has picked Windows as the preferred platform for running Photoshop, After Effects, and Illustrator. I don't know how many Mac people this will upset, but given the large hold Apple has on design pros and film, this seems like a bad move on Adobe's part."
it just shows graphs that say the PC is faster than the mac doing stuff in after effects..and at the end it says, "While the computers used in this study are no longer the fastest in their respective classes, the information is still valid"
------ Work is so much easier when you don't
The images appear to be incorrect.
If you look at the first image, it has two times, 54 seconds and 1 minute 25 seconds. The second time is shown at well over double the length of the first, even though it only took ~50% longer. If you look closely, you will see that 1:25 got placed at 1.25, and 0:54 got placed at 0.54, hence the error.
Any of the images where the minutes are different are going to be skewed a fair amount. The error will decrease as the minute difference increases.
Well, it seems what Adobe has actually done is a bit less inflammatory than what the headline suggests. On the hyperlinked page, they simply display the results from a performance benchmark that indicate a 3.06Ghz P4 outperforms a dual 1.25Ghz G4 by a wide margin on some tests, which is a little confusing as the article said the P4 was a 2.53Ghz. Whatever.
This changes very little and seems hardly worth the effort sensationalizing.
Photoshop and Illustrator are the most annoying programs to use under Windows. They both use the MDI (Multiple document interface) model for drawing their windows which makes it very difficult to utilize the avalible screen space.
MacOS and even the UNIX versions of Photoshop/Illustrator do not suffer from the same design flaws.
The clockspeeds of the computers mentioned in the introductory paragraph on that page don't match up with the clockspeeds of the computers in the charts. I'm wondering what other errors are present as well.
Also, this doesn't look like an Adobe recommendation so much as Adobe showing one group's results of a comparative test. There is more to a computer than render speed, just as there is more to a computer than compile speed.
mbbac
The link doesn't say anything about Adobe preferring one platform over another, in the slightest. It's just some graphs indicating that PCs as a class perform better than macintoshes, which is something that i don't think anyone is denying at this point.
While that kind of does seem like an endorsement of the PC on adobe's part, it also is just good business sense to explain to your customers what hardware your software runs best on.
Speed at raw data-crunching is just one of the factors in which computing platform you are going to use, though if you're using AfterEffects or Photoshop or something it's going to be a much, much larger factor.
Irritable, left-wing and possibly humorous bumper stickers and t-shirts
If there ever was a definition of lies in statistics, that's it.
I am in full agreement that a high-end wintel box will out-render any mac, and I run a mac-oriented media production unit. But...
I read said article in a copy of some trade mag. I was annoyed and immediately recycled it, because:
1. Focus on hardware alone misses the point and causes people like me extra unnecessary work.
2. The 'article' in question is being used as a large advertorial insert partnered by adobe and intel, IN DISGUISE.
3. The 'article' and all the surrounding 'information' leaves out crucial issues like uptime, human interface, Return On Investment, training time, technical support, h/w-s/w integration, and other minor economic factors that add up significantly over a year's production. These factors more than make up for the difference in render time. [Ask any of my friends who've tried and tried to make Premiere do what it promises how much time they've saved...]
I keep telling my interns [and anyone else who will listen]: it doesn't matter how fast your hardware is if your wetware is lagging. Speed requires optimization all the way down the signal path, starting with ideas.
Damn those pesky terrorists
Adobe Carbon code running on Mac OS X has nothing more in common with Unix than it did when it was running on Mac OS 9. Carbon is a subset of the old Toolbox APIs and really is about as far away from porting something to Unix as you can get. Sorry.
I doubt Adobe is the actual source of the stupid axis-labeling error -- Adobe attributes the images to Digital Media Net, the parent of the site that published the article this is based on -- so I'd imagine it more likely that the error's on DMN's side.
It's interesting to look at Digital Video Editing, the site that published the original article entitled "Macs vs. PCs III: Macs Slaughtered Again".
I'm not enough of an editing guru to comment on the validity of the tests, but the writing is strikingly unprofessional: "Mac stalwarts will cling to the notion that Mac OS X is so much better and easier to use than Windows XP". He's obviously got an axe to grind. Writers who compare Macs and PCs and *start out* with a chip on their shoulder kind of piss me off.
It seems quite possible that Adobe asked the author for a couple of images, and he came up with these worthless, mis-scaled pieces of junk to force his own point. But maybe it was an accident, and I'm just a pessimist.
Yes. However, Adobe never released a more recent version, and won't even sell you a license for the old IRIX Photoshop, despite the fact that SGI still has a page up for it.
Well, as I understand it, it went something like this:
Adobe had a great team of programmers working on Premiere once upon a time. They created Premiere 4 which was a staple of low-cost video editing (and where many of us got our start). It was a quirky program but very flexible and people liked it.
This team had a radical new idea for the next version of Premiere, but for whatever reason it didn't work out with Adobe. That team, possibly in its entirety, left Adobe, as a purchase by Macromedia.
Macromedia continued to work on what was now known as Final Cut and Adobe hired a new team which have birth to the abortion (my opinion) that was Premiere 5.
Apple then cut some sort of deal with Macromedia (possibly when Avid were being dicks) and picked up Final Cut.
Rest is history. You could see how Adobe is sore. They would never have given Apple the Final Cut project but somehow it ended up being 'laundered' through Macromedia.
If anyone has any corrections to this feel free to reply.
If Jesus wants me it knows where to find me.
Vote Quimby.
Actually, this is far more likely to cause Apple to start working on The Gimp and adding the features that Photoshop has that The Gimp lacks.
And GM will switch to hovercraft designs because Goodyear said nice things about Ford.
The GIMP is no competitor to Photoshop. Sure, it's an impressive piece of free software, but graphics professionals are not going to switch from Photoshop. On average, each user has years of experience, probably has taken courses in using Photoshop's advanced features, and may have a considerable investment in plug-ins on which they rely to do their job. They would sooner switch personal computers before they would switch image processing software.
Allow them to add features that Adobe does not see fit to add
Apple's software developers don't work for free. The GIMP is under the GPL which means that Apple would have to give away all of the code that they develop for it. They would not be able to sell it (they could try -- and become the next Mandrake). Thus, they would be paying software developers to add features to a Free Software program so that Apple could have a substandard Photoshop replacement that would still run faster on PCs. That's hardly going to result in massive sales of Macs.
In short, this could be a good thing for Free Software.
This has nothing to do with Free Software. There are only three possible outcomes for this and neither of them involve Free Software:
1. Graphics designers switch to PCs.
2. Apple switches to x86 architecture CPUs.
Slashdot strikes again. What Adobe did: link to that pc/mac comparison that one guy did that showed a Dell 3Ghz outperforming a Mac dual G4 1.24 Ghz, under the title "Prefer a PC for DV?" The link is in the "digital video products" page -- Premiere and After Effects. (Fair enough too -- if you were doing heavy video editing, it's a useful reminder of performance.) It is *not* under any of the other product pages that I can see.
So how does this justify the slashdot interpretation: "Abobe has picked Windows as the preferred platform for running Photoshop, After Effects, and Illustrator"? I don't see it.
A.
Has anyone noticed how screwed-up those graphs are, especially the first one? It says the PC took 54 seconds and that the Mac took 1 minute and 25 seconds. If you measure them both in seconds, then that is 54 seconds versus 85 seconds, but the Mac bar on the graph is more than twice as long as the PC bar.
.54 according to the lower index. Is this index supposed to represent seconds or minutes? If it's seconds, as suggested by the fact that the PC bar lines up with .54, then why is there a marker at 0.9? And more importantly, why does the Mac bar line up with 1.25, and not .85?
The PC bar lines up with
Tricky! (But not as tricky as the incredibly misleading title on this SlashDot posting.)
Mike van Lammeren
It will challenge your head, your brain, and your mind.
I'm making an exception this time because I can save a lot of people time before they delve into the over 500 posts of reaction to this story (at least for those who read /. in threaded mode).
Adobe is not expressing a preference for Windows PC's
The linked page is called "pcprefered.html" because it is the page which is brought up on the Adobe Digital Video Products page when you click on a link that says: "Prefer a PC for DV? See what an industry expert has to say about PC vs. Mac for video editing."
In other words, those who followed the link from The Adobe DV Products Page are indicating a preference for PC's. Since it's a page for those who prefer PC's, it's called "pcprefered.html".
There is nothing in the body of the page to indicate that Adobe has any preference for PC's, reccomends PC's over Macs, or even likes PC's. The page is a mirror of some Dell vs. Mac speed tests that some guy did. That is all.
By deep-linking to this page out of context, the person who submitted this was obviously trolling... perhaps hoping that the article would not go up until a little closer to April 1.
You may now safely ignore all of the responses below and move on with your life. No need to mod up this post, I'm already posting it at 2. Save your mod points for a real article.
Information wants to be anthropomorphized.
maybe it was changed since then, but the 3:47 time bar in that image falls between 3.5 and 4.0, almost exactly inbetween. 3.75 seems like a logical measurement of 3:47, don't you think?
Nope. It runs in both Mac OS X and Mac OS 9. It's a Carbon app through and through. Doing it in Cocoa would have required a complete rewrite.
I'd mod myself up to get this read, but oh well....
This computer cretin was simply out of place while making this review, so bear it no weight. In the following blocks, I provide sufficient information to prove that the "tester's," Mr. Charles White's, review neglects important information, is not thoughtfully written, and is irresponsible.
First, notice the title in the article. It is definitely biased based on his word choice. He is obviously a pro-pc "I hate mac" person considering all the time he spends describing the PC. Ars Technica does a much better job at keeping tests objective and posting all tests completed. The tester probably left out a few tests for the sake of his article, adding to his irresponsibility.
Major errors in this article, including, but not limited to....
1. On page four, he indicates that the graphics card was a newer ATI Fire card (a back of the pack pro graphics card, NOT a consumer one as he claims. The same NVIDIA card in the apple can be put into the pc, but was not. Dell gives buyers several options, and he could have picked the same card for both.), and the mac had a consumer grade NVIDIA whatever Ti with 128 VRAM.
2. The Apple hard drive the tester used has only 2MB of onboard cache, while (rather cutely) the hard drive tester substituted for the PC has 8MB onboard cache.
3. Tester talks about hyper-threading. He obviously has not read the documentation Intel provides, as I have, because he mentions:
Without boring you to tears, I?ll tell you that hyperthreading is a new technology from Intel that makes one processor act like two. It doesn?t double the speed of a processor, but makes it able to do most operations faster, and is particularly effective if you?re doing more than one thing at a time with your computer (multitasking).
No, he is incorrect. Hyperthreading on average makes the processor 30% more efficient on a Xeon processor. That percentage drops a lot when one adds as many pipeline stages as the P4. I would estimate that he got less than 10% in performance gains with his hyperthreading in a P4.
Intel does not provide lots of Hyper-threading on a P4, not because of patent issues, as the reviewer claims, but because it is simply inefficient. His own two tests with other computers and with the Dell 350, hyperthreading off, make that clearly evident. Intel itself would rather have someone buy Xeon processors, for they are much more efficient.
4. He claims a bunch of other nonsense about how certain hardware (RAM, Logic board throughput at the processor bridge) makes the PC faster, which is not the case. In fact, much of what he talks about is totally unrelated to the specifics of his tests.
5. Several of his tests rely more on the graphics card and less on the processor. That skews results massively since tester uses a Fire-class card. He should have gone with a 3DLabs wildcat 4 if he really wanted to differentiate the scores.
6. Again, vector operations being performed in all these tests are not the same as floating point operations. Few of his tests used the floating-point abilities of AltiVec. The tester is a "hardware kiddie" if such a thing does exist, and there is a difference between a processor and a graphics card.
7. Many of the tests he posted were tailored for the PC. The tester probably did other ones, but the Mac must have done a decent job on those. How about opening (oops, 8MB hard drive cache), a 150MB Photoshop file in 2400dpi resolution. That file must contain anti-aliased text and a few high-resolution photos.
Finally, I would say from my experience at building workstation hardware and writing reviews for other hardware, that Mr. Charlie White has extremely limited knowledge, provides much "bs" to fill the article, and that he is unqualified for making his review. The site he posted on was digitalvideoediting.com. It is out of the scope of his review to even touch on Photoshop, other 2D, or non-digital video sources. Mr. White has neither the knowledge, expertise- check his credentials on the primary source article -nor the objectivity to make this review. His article is simply irresponsible journalism.