Design/Production filter: Going from Mac to PC.

November 29, 2005 7:33 AM

What concerns are there when switching from an all Macintosh (Quark 4.1, Pshop 7, Illustrater 10) shop to an all Windows shop running new PCs and the Creative Suite 2?

Disclaimer: Please, no Mac vs PC religious wars. I'm looking for pure facts, i.e. information that can solve this problem and make this as change as smooth as possible if need be.

Background: We're the production department of a weekly newspaper, we build ads and layout editorial pages. We send the finished product via PDFs to the printer.

1. How do we transfer our large library of fonts, which are a mix of TrueTYpe and Postscript?

2. On the PC side should we stick to Postscript fonts or does it matter?

3. I know Indesign will open Quark 4 files ok (they'll need to be tweaked I know), but is this viable for a massive conversioin, say 3,000 files? Is there a utility that will do this?

4. Is there something equal to Applescript on the PC side?


5. Our server is currently running OS X server 10.2.8, what should we switch to?

6. What sort of font management software should we use on the PC side?

7. What type of workstation is good for Design and Production work?

8. What type of computer is good for a server?

9. What type of monitor for color correction?

10. Is it true that Creative Suite 2 is cheaper for the PC?

11. What other concerns/problems might there be?

12. Working strictly from a cost view, do you think (not feel, but THINK) this conversion a good idea, with the goal of making the production department more cost efficient (since PCs are cheaper) and not wanting to go through Apple's conversion/problems to Wintel processors? If so, why or why not? Supporting evidence/links (either way) is greatly appreciated.

posted by Brandon Blatcher to computers & internet (30 comments total)

Regarding the Quark-to-InDesign conversion issue, be aware that conversion is never perfect --you will ALWAYS have to tweak the files by hand. So.... from a cost-savings benefit, are you willing to pay for the extra (my estimate based on your qty of files to convert) 500-1000 hrs of designer's time (probably spread over the span of a year, but nevertheless a factor) to facilitate your conversion?

Also, does anyone in your shop surf the web? If so, the extra IT dept. time for virus-prevention and maintenance you will need must also be factored into the conversion cost.
posted by Chrischris at 7:51 AM on November 29, 2005

I don't know the answer to all of these questions, because frankly I would not consider making such a huge transition for the reasons you list.

Are your macs currently severely outdated and in need of upgrades? Unless they are, you can neatly sidestep the intel transition by just not upgrading to new machines until the intel lines are into their second generation and complete software support is available without an emulater.

I've always thought the cost difference between PCs and Macs is overstated, especially in terms of the professional grade machines needed to run a print operation. For instance this review from computerworld states that getting a PC with similar performance to the new G5 quad processor Macs would cost double the money. I'm not really sure you would be saving enough money to justify all of this work.

Finally, I can give you a simple answer to question 10. At least according to Amazon.com, CS2 software costs the exact same for both platforms.

I would fear that your concerns are just the tip of the iceberg and that a tremendous amount of time and resources (and initially, money) would be spent resulting in marginal, if any, cost saves, and decreased performance over time.
posted by jasonlatshaw at 8:04 AM on November 29, 2005

12. I think it's a bad idea. While *some* COTS PCs may be cheaper up-front, the total cost of keeping a Windows machine running, with AV, spy-ware removal, lost HCI productivity would make this a dollar loser in the long run. What more, on a feature-to-feature basis, the Apple machines are equivalent or less expensive. (Source "Linux News")
posted by unixrat at 8:05 AM on November 29, 2005

I'm the IT admin for an Art/Pre-Press Dept of about 30 Macs (G5's, G4's) a few Xserves and xraids (about 12tb) and about 20 NT servers (platesetters, digital presses, RIPs, etc). We looked at moving from dual G5's to Dual intel Dells or IBMs. We setup a new PC with all of the apps that our G5's have. We killed the PC in a few days. It couldn't handle what was thrown at it. We ditched our 9tb Windows-based SAN for a 12tb xSan and haven't looked back.

1. Use something like transtype to transfer your fonts. This will be one of your biggest issues.

2. I don't think it matters.

3. Not that I know of.

4. Get someone who can script out C++

5. If you're ditching macs entirely, any PC server should work.

6. Suitcase. that's about it. and it sucks.

7. You're gonna look at spending about $2-300 more per box if you go the Dell/IBM workstation route. We looked at this, looking at getting Workstations (not Desktops) to replace our G5's. A few boxes and this isn't a problem, but with the 30 or so that I manage, this drove the cost wayyyyy up. Don't skimp, pay for a quality box with a warranty. Don't white box a system, I'm serious about this. You will regret it.

8. HP has been good for us, Dell's too.

9. We run the blue LaCee (sp?) we also have a lot of Apple Cinema displays. We do a lot of color correction. We also have some nice Sony's that our color guys absolutely love.

10. I don't believe it is. Talk to a group like CDW about getting a volume license. One set of media, one license number.

11. If you're still gonna run a mixed environment, don't use the Services for Mac on the file server. get something like ExtremeZ-IP. We had a hell of a time until we begged the people who make EZ-IP to let us be an alpha site.

12. I BELIEVE this is not a good idea. See #7. We've looked at this, but if you have long-time Mac users. Your production will suffer GREATLY. If you are worried about price, Apple's Enterprise division will help you a lot. You can get a beefed up machine with Applecare for about what you would pay the Apple store for a stock machine with a standard warranty.
posted by mkelley at 8:18 AM on November 29, 2005

I don't think it's a wise decision either, cost-wise. You CAN convert all your fronts from Mac to PC using TransType or FontLab, but a converted font has the chance to lose important kerning and hinting details in the process. A converted font is never as trustworthy as one made specifically for its platform. You'd be wiser to re-purchase your font library for the PC, but that would be ridiculously expensive.

Honestly, I have only used TransType occasionally for display faces, and not for any fonts used for extensive body copy. The latest version of the software may do a better job than I am giving it credit for, but I, myself, would be wary of using a font library that was entirely converted from another format.
posted by Robot Johnny at 8:24 AM on November 29, 2005

4. There is something called Windows Script, but it's a bit more low-level and less user-friendly than Applescript, IMHO. Here's a pretty good comparison of the two.

5. Microsoft's equivalent server product would probably be Windows Small Business Server 2003. Depends what you're using the server for, though. File storage, email...? You may be able to get away with something simpler and less expensive.

11. You'll definitely have some wasted time and frustration while you get used to the Windows interface, and its quirks.

12. Unless you by new computers every six months or something, I strongly suspect that any saving you make on PC vs. Mac hardware will be cancelled out by the time wasted learning the new system, converting fonts/resources, dealing with unfamiliar technical problems, buying new software licenses, etc..

And for what it's worth, I also reckon that the Apple PowerPC to Intel transition will be a lot more seamless than a Mac OS X to Windows transition. After all, it's the operating system that you really interact with, not the hardware.
posted by chrismear at 8:27 AM on November 29, 2005

I would Think this conversion makes no sense at this time. It will not make the production department more cost effective because

- you will need new licences for many of the fonts (If you are doing everything above board that is).

- Converting all your old Quark files will be a major PITA. I'd agree with ChrisChris - it will take 100's of hours of a designers time. Same could be said if you are sticking with macs and going the ID route though. This is a guess but I'd say Mac Quark - Mac CS will be less painful than PC Quark - Mac CS.

- You will need anti-virus software and probably a dedicated Tech guy. This is fact, not speculation. Macs are known to be more productive with less issues in a Print shop.

- get an Electron Blue La Cie as your monitors - brilliant.

- If the staff are used to macs they will possibly not thank you for the change and there will be a performance hit as they get used to the new way of working.

I honestly can't think of one benefit to this conversion at all. The boxes will be cheaper but in the long run you will be paying a fortune.
posted by twistedonion at 8:32 AM on November 29, 2005

I believe my e-mail is on my profile, if you have any questions.

I agree with Chrismear on everything except #5....go Win Server advanced and load up as much memory as you can. Use that server as your file server and nothing more. Get another box for Exchange...or use your old Macs as your mail servers.
posted by mkelley at 8:33 AM on November 29, 2005

sorry, PC Quark - Mac CS should read Mac Quark - PC CS

Also, you mention Apples conversion problems... if you made it from OS 9 - X with few problems you will find PowerPC - Intel will be a breeze.
posted by twistedonion at 8:35 AM on November 29, 2005

Thanks for all the responses so far, it's given me a lot to chew over.

A few answers to questions:

Are macs currently are old (original imacs, G3, 400mhz G4).

We're currently using the files server for storage, with separate user accounts for different departments.

More questions:

Converting fonts using Trans Type: Has anyone done this on a massive scale? We have a lot of unique fonts (we're an A &E paper) which I worry about.

Anyone have any specs on the alt weekly industry or newspaper industry use of PC vs Mac in the graphics/design department?

On the PC, we'd be using Firefox and Exchange for browsing and email. We're sitting behind a nice firewall. Are there any other security problems to be aware of?

My current idea on this is to use Indesign to build the ads for a few weeks, then switch the editorial people/layout duties to Indesign. That way there's some people who are familiar enough with the quirks to help those who aren't and we would have seen most problems. Anyone see any problem with this?

Does anyone have any links to departments that have SUCESSFULLY done this?

Can anyone recommend SPECIFIC PC workstations? We use Photoshop heavily, so we need speed, something that can match the two 2ghz double processor machines we currently have.
posted by Brandon Blatcher at 8:48 AM on November 29, 2005

You'd probably be better off building new templates and masterpages in InDesign instead of converting so much. A newspaper's layout is very standardized and it wouldn't be hard.
posted by amberglow at 9:02 AM on November 29, 2005

BTW, how many machines and people are we talking about here? Just to get a better feel for the cost/time benefit... weigh-up... thing.
posted by chrismear at 9:08 AM on November 29, 2005

Brandon Blatcher: 12. Working strictly from a cost view, do you think (not feel, but THINK) this conversion a good idea, with the goal of making the production department more cost efficient (since PCs are cheaper) and not wanting to go through Apple's conversion/problems to Wintel processors? If so, why or why not? Supporting evidence/links (either way) is greatly appreciated.

Strictly from a cost view, hardware costs are going to be dwarfed by human costs. Managment of computer users are going to matter a lot more than mangement of information systems.

Just from a support perspective, I think the switch to Intel for OSX is going to cause fewer problems than the switch from Windows XP to Windows Vista. BareBones and Mathematica have already announced successful ports, while Adobe appears to be developing its future products for OSX/Intel. With most of the same software running under both OSX/IBM and OSX/Intel, the support and training problems will be minimized.

The XP to Vista switch strikes me as being a potential bigger speedbump as someone who works in mixed-compting environments. With Microsoft, you are looking at switching now, and switching again in another 12-36 months.
posted by KirkJobSluder at 9:09 AM on November 29, 2005

I'd recommend the CS/CS2 upgrade, but not necessarily the platform change to Windows. Many of your troubles will be caused by the move to Windows, while relatively few will be caused by the switch to CS. You'll also find that Windows gives you very little that Macs don't already, while CS/CS2 is a huge step up from Quark.

Font conversion is iffy at best and my experiences with converting Mac Type 1s to Windows Type 1s is that the printer font never works. To do it properly you'd have to get Opentype versions of all your fonts, which doesn't really sound like a viable option if you're trying to save costs (plus I don't know if Opentype is widespread in the printing industry yet). Depending on how large your outfit is, you're going to have to get new tech people or retrain the ones you have, not to mention all the editorial/production staff; even in an alt-weekly situation with fewer computers, you're absolutely going to run into problems you haven't seen before with Macs. This is not a "Mac is better" observation; I'd say the same thing if you were already on Windows and were moving to Macs.

Plus, as mentioned above, corporate-level PCs from brand-name manufacturers are going to cost only slightly less than the Apple equivalents; at that level you're paying for the (hopefully) awesome tech support, not the components. Finally, the switch to Intel on the Mac side shouldn't be a big deal; the announcement earlier this year was really a glorified developer announcement, and theoretically shouldn't have any impact on the end user. In other words, if you're worried about legacy support for G4/G5 systems, don't; those systems will likely be supported for a while yet (and besides, you've managed to keep Quark 4.1 this long, and it's a freakin' dinosaur). If you're worried about purchasing Intel-based Macs that don't work, I wouldn't be too worried about that either; getting CS/Quark to run properly on Rosetta is, I'm sure, a top priority for the Apple devs.

Regarding Applescript: the problem with most Windows equivalents is that most software packages don't play with Windows Scripting Host, etc. straight out of the box. So if you've got a lot of Quark Applescripts you wanted to keep for use in, say, Quark 6 for Windows, you'd be out of luck. However, because you're planning to move to CS, you'd be in luck; all the CS apps are scriptable using Javascript/Visual Basic on Windows.
posted by chrominance at 9:10 AM on November 29, 2005

We're talking five people, six machines.
posted by Brandon Blatcher at 9:12 AM on November 29, 2005

I held onto Quark 4.1 for so long 'cause it's only version that Indesign will open and I knew we would eventually switch to Indesign.
posted by Brandon Blatcher at 9:14 AM on November 29, 2005

So, given that you're planning on moving to InDesign regardless of platform, the only reason for this switch is to save money on PC hardware, right?

Let's say that you save $1000 per machine by buying PC instead of Apple. Even if you upgraded all your machines every two years, you're only saving $3000/year. If you upgrade less frequently (and given that you've still got G3's in use, I'm guessing this is more realistic), then your annual savings on hardware will be even less.

(I specced up a Dell Optiplex and a Dell Precision to 3.0 GHz, 1 GB RAM, 160 GB HD -- you're looking at $1200 - $1800. Apple dual-core 2 GHz G5, same RAM and HD, about $2100.)

What with all the font converting/repurchasing, wasted time while getting used to the new OS, etc., full licenses for Photoshop rather than just upgrades, additional technical support etc... realistically your first year costs for the switch are likely to be at least $10,000.

(Wasted time: say, 5 minutes a day on average for each of you; that's 100 hours in the year, and so a few thousand dollars. If you have to repurchase a couple of dozen fonts, you could be looking at four figures again. Full version of CS2 is about $450 more than an upgrade version x 6 machines = a few grand.)

It's very rough and ready, and you'll obviously have to do this with your own figures, but the end of the day, this is what it's all about. If recouping your costs in three years and then saving a few grand per year is feasible and worthwhile to your department, then it's a good idea. Otherwise it's not.
posted by chrismear at 9:55 AM on November 29, 2005

Can you do a gradual transition rather than one that's so abrupt?
posted by MegoSteve at 9:59 AM on November 29, 2005

Also, I believe that using something like Transtype to convert fonts from one platform to another is illegal except if you are using freeware faces. Fonts are like any other piece of commercial software. If you'd never dream of using pirated apps, then please don't pirate typefaces.

Just another reason why this proposed move from Macs to PCs is completely nuts, IMO.
posted by ninthart at 10:47 AM on November 29, 2005

Random practical question:

How do you open a Mac Quark 4.1 file in Indesign CS2? Does it need the .qxd extension? Drag and Drop? Open from within Indesign?
posted by Brandon Blatcher at 11:56 AM on November 29, 2005

MegoSteve, not sure what you mean.
posted by Brandon Blatcher at 11:56 AM on November 29, 2005

12. I think it's a dreadful idea. You don't know shit about PCs, judging by the questions you're asking, so you have no idea of the horror speedbumps ahead. Drop a few hundred on a Dell as a proof of concept and work with it for a week. You'll drop the idea sharpish. Or your designers will, if you let them near it.

I know of editorial teams gearing up for intra-company war over a move such as this. The Mac has a huge base in publishing for a reason. The price premium is vastly over-stated, and the TCO for Apple is much lower, especially in support terms.
posted by bonaldi at 12:29 PM on November 29, 2005

I think it's a dreadful idea. You don't know shit about PCs, judging by the questions you're asking, so you have no idea of the horror speedbumps ahead.


It's a company wide change. We'll have a co. support specialist to help us through the changes and as they've already change one, smaller paper, over it shouldn't be that big of a deal, right?
posted by Brandon Blatcher at 12:43 PM on November 29, 2005

Wrong. If these specialists were helping you, they'd have answered all those questions already. they're going to leave you to swing.

The only reason companies switch to PCs is they (incorrectly) think they'll save cash, or because of vested interests in IT who only know PCs and couldn't give a fuck what's better for the user. If they're saving cash, they're not going to spend a lot holding your hand on the move. Resist man, resist!
posted by bonaldi at 1:37 PM on November 29, 2005

This sounds like someone who has the CEO's ear got a brain fart and now the CEO thinks it smells like roses.

Speaking frankly, and as someone with about 15 years of experience working in IT or related fields, the vast majority on PCs: It's a really bad idea.

As KJS has pointed out and lots of other folks have implied, the human costs will dwarf the license and hardware costs. And the unforseen license costs alone (font licenses, duplicate software licenses -- Macrodobe isn't going to let you just translate those licenses, you know -- additional unforseen software like antivirus and software firewalls, etc) are pretty significant.

You can definitely make Windows work for a graphics and print environment, and you could probably save money at it. But it would take a lot of work to save that money -- you'd want to be looking closely at your opportunity costs -- and I doubt you could save any money at all by migrating.

Also, the PPC to Intel shift probably isn't going to cause much problem for you. You'll want to hold off refreshing the workstations until you have all the apps you need on Intel, but your servers I expect you could migrate as necessary. Remember, Apple has been doing proof-of-concept builds on Intel since 10.0.0. They know how to make it work for the Apple stuff, which is what matters most on the servers.
posted by lodurr at 1:37 PM on November 29, 2005

Here's a thought to take to your bosses: If someone is arguing that the migration will save money, make them show the model. Make them show it in real time. Change the assumptions and see what happens.

If they haven't got a model (e.g., in Excel) that lets you change assumptions and see the results, then they probably don't have a sound argument for cost savings.
posted by lodurr at 1:40 PM on November 29, 2005

It's a company wide change.

Why? What benefit is supposed to accrue from adopting a single platform, or indeed, moving from one platform to another?

We'll have a co. support specialist to help us through the changes

What kind of company? Are they the ones making the reccomendation? Are they the ones selling you the PCs? Do they have any companies they've "helped" which still use Macs?

People often assume that one platform is better than two. But that's generally what it is, an assumption, not a tested theory. Or, being cynical, it's really just a decision which makes life easier for them, because they don't do Mac support.

they've already change one, smaller paper, over it shouldn't be that big of a deal, right?

Have you talked to the people at that paper?

In case you can't tell, I'm one of the people who thinks it's a terrible idea.
posted by AmbroseChapel at 3:40 PM on November 29, 2005

Brandon, do your bosses, IT folks, bean counters know that there are plenty of companies that operate just fine using both platforms? Especially in publishing, it's quite common. And if it seems counter-intuitive, it's because it works. For example, your advertising and circ people may need (or strongly prefer) Windows for the software, the creative and editorial types Macs for the same reason.
posted by rob511 at 3:59 PM on November 29, 2005

Clarifications:

Yes, *I* think it's a bad idea.

But I'm trying to fair and look at this objectively and how to make it work. I'm also trying to demonstrate that this is a bad decision in a non combative, friendly corporate way, that referrers to the bottom line.

So I'm writing up something to present to the higher ups, while trying to come up with answers/alternatives to any misgivings they might have.

Ambrose:
Again, it's perceieved as being cheaper and the IT department said so. The support specialist is part of the IT department and was formerly a production manager for one of the satellite offices of the company.

We're a newspaper publishing company and the IT/Tech department is advising on this. Yes, they're PC based and want everything to be PC based as it's easier (cheaper) to maintain in their eyes.
posted by Brandon Blatcher at 4:09 PM on November 29, 2005

rob511 has a major point: There's no good reason that Macs and PCs can't interoperate perfectly well, especially in that kind of environment.

No, they can't manage the Macs with NT domain management tools. But for substantially less than will be lost doing the conversion, they could afford to hire a Unix networking consultant to resolve any actual integration issues they have.

Again, if they want to claim it's cheaper, they should have a model that shows it's so. If they've produced a proposal or a project plan, it should have enough numbers in it that you can reproduce their model. If it doesn't, then there's really no case for the switch.

If they want to make it about cost savings, they should be able to get specific about them.

Good luck. I know the good fight can be hard to fight. Also remember that at the end of the day, lives (probably) won't be lost, either way.
posted by lodurr at 5:04 PM on November 29, 2005