Hacker News Reader

Comment by jphoward | Hacker News Reader

I wonder who the target market is, though? The iMac Pro can cost £5000, so I assume more performance and hence price than that? The only users who I can imagine willing to justify paying such a premium are those who feel they must remain at the cutting edge of performance and are willing to pay for it, but surely they will have left the Apple ecosystem by then?

It's as if Apple are saying "we're making a computer for people who demand peak performance, but are willing to use cheap outdated hardware until we get round to it."

View on HN

If you do any iOS work you need macs as build machines because of Apple's lock in. Spending the money on an iMac is just a waste of a monitor and if you're already spending the money on the hardware you might as well get something that can do render jobs as well. You need GPUs for that.

The Mac Pro doesn't include a monitor, so I can see it starting at less than the iMac Pro.

err no not if they do it right they will be offering the top of the line Xeons and one or two hefty GPU cards.

The new imac pro has the entry Level xenon and still throttles after 10 min.

They might offer that as a configuration option but not the base config.

I'd assume mostly people who want really hefty GPUs (and for whom eGPUs aren't an option). The iMac Pro won't help you there.

In the article it sounded as if they actually intend to achieve modularity fort the Mac Pro with eGPUs, or peripheral hardware in general. A mistake in my opinion.

> I wonder who the target market is, though?

Schools for sure. The staff believes that Macs are better and have bough trash cans to replace the cheese graters.

I think this is where Hackintosh comes in. You can buy top of the line, macOS-compatible hardware for 1/3 of the Apple premium.

It takes effort to set up, but not that much.

To me the value of a Mac is not macOS. It’s having a complete device solution that just works.

I don’t want to fight with my workstation all the time. I want to use it to do... work.

I have been battling Apple about my 2013 mac pro for a year. You have to be careful about the monitors you buy because Apple will say it's not their problem when your monitor knocks your wifi offline (personally) or some weird shit. Personally, I think for anyone deliberating between an iMac Pro and a Mac Pro, you should go with the iMac Pro because Apple has always been an all-in-one company--it's one of the few things they are still good at these days. I used to work on the older mac pros in high school and never had problems, but those days are over. If I knew hardware well enough to fix it I would have gone Hackintosh.

I used to have a '14 RMBP with the Apple display and it was wonderful. Now I have a '17 13" MBP and it only has two USB ports so I have a third party dock and monitors because Apple doesn't offer a solution. The entire experience is awful but still more reliable than any Windows laptop I have ever run, although the gap is closing.

Hackintosh doesn't work for me because my business isn't fixing laptops. Apple used to provide a full solution and it was really great.

Hackintosh run very stable these days even more so now with Clover. But you need to have fun at doing that

Yeah I get it, I even consider that kind of thing fun in my spare time although I have other hobbies these days. In a professional (corporate) environment I don't have time to fix my laptop. I'm lucky to work in a place that has an outstanding IT department who provides what we need (including outstanding Macintosh support) but a Hackintosh is not on the list.

I am also sure the legal department would shut down a hackintosh real quick.

You can't knock everything else just because it's not a Mac. I've personally had my fair share of Mac problems and when I worked Help Desk we had our fair share of Mac Laptop problems that were comparable if not greater than our failure rate of Thinkpad's. This was in 2017 too.

I’m not knocking it because it’s not a Mac. I’m saying when Apple sells a complete solution it is generally trouble free.

That may have been the case 5 years ago, but it certainly isn't now. I work in a mostly mac office and not a single one of my mac fanboy coworkers has a single nice thing to say about the 2017 macbook pros.

Right because it's just a laptop and you have to buy a bunch of other stuff from third parties to go with it. I explained this in a sibling comment.

In 2014 Apple would sell you a monitor that doubled as a dock which worked great with the RMBP and Apple's own wireless KB and Mouse and even Time Machine.

Today you're stuck with USB-C docks which are pretty awful. Pretty much on par with a Dell or HP laptop dock. The new Macbook Pros are the worst device I have used from Apple.

The thread you are replying to is about hackintoshes. The point of a Mac for me is that it Just Works (tm). If I want to fight with something I'll buy a Thinkpad and install Ubuntu. I have no interest in fighting with macOS, it's only useful to me if it is easy.

> Right because it's just a laptop and you have to buy a bunch of other stuff from third parties to go with it.

No. Their complaints are about the laptop itself. The keys are constantly breaking. The software constantly crashes. We're all a bunch of consulting devs and the 2017 macbook pros have cost every single one of them at least a week's worth of billable hours in the past 6 months. Most of them have gone back to their 2015 models.

Perhaps this is part of the reason why Apple is moving away from Intel chips to some of their own creations?

Statistically there are almost no hackintoshes, it would be a very dumb reason to switch from Intel chips. The much more likely reason is that Intel chips have hit something of a performance wall and Apple also runs a much more popular ARM based platform.

More likely still is that one chip set convergence lowers the cost of OS development by combining both platforms into one.

I simply cannot imagine running something like that in a professional context. For me, "maybe this next update will install correctly" is a non-starter.

For consumers who just really want to run macOS, I can see the appeal.

I agree, in an educational institution or large corporation, no way. But in a small shop that lives and dies on productivity, where one can't (won't?) migrate away from MacOS, a Hackintosh may be the only option available.

It's the market filling in the gaps of Apple's engineering & marketing plan.

> in a small shop that lives and dies on productivity

A computer that takes a bunch of fiddling to set up and might not work if updated doesn't seem like a productivity booster to me.

IT does the fiddling. Creators gain performance and productivity.

You're a small shop. Do you even have the funds for IT, or is it just one of the hats that your main system admin or reliability engineer puts on?

That depends how much your creators' time is worth, i.e. the return on increased performance.

If Hackintoshes have motivated Apple to release a new Mac Pro after a multi-year gap, then Apple-only shops will benefit from the Hackintosh investments of other shops.

Yes, that's my point. How much money are you saving? Maybe a couple thousand dollars? Balance this with the time you'd need to spend on setting this up and getting it to work right–a couple days, a week? It's not obvious that this would always work out, especially so if you don't have in-house IT and would either have to get a creator to learn how to do something like this or hire someone that does.

Right, so those people keep using Apple and wait N years for a new computer.

Other people use a Hackintosh and help to define requirements for Apple's next computer.

Apple gets free R&D. Those who need performance get it. Everyone eventualy gets the benefits in the next computer.

I agree.

Works perfectly in a home office and saving $2500 is not a small thing.

Not all updates are critical.

This is a myth that keeps going and going.

A Core i7 is not the same as a Xeon. ECC memory matters. The Apple premium simply isn't that much higher.

I also wanted to add that the cost of a person's time eclipses the cost of hardware (especially with computers). As a business, I simply wouldn't risk having days of downtime from a bad update, just to save $2,000 for something that's used for years, when the lost time in paying a creative person's salary will eclipse that.

And everytime, someone mentions how ecc memory is important. But it's just not. There is a performance penalty, and the real use case for ecc memory is vanishingly small for a workstation.

What could you possibly be doing on a workstation that requires ecc?

For every 8 gb of memory, you are potentially looking at a single bit flip every 9 years. With ecc, only once every 45 years. And in exchange, you are looking at a 10% performance penalty.

What could you possibly be doing on a Mac pro that would need that kind of accuracy?

The error rate is significantly higher than that. ECC also makes RowHammer much harder to pull off.

ECC for security on a desktop to combat RowHammer?

Even if it is higher, what are you doing on a workstation that would benefit from lower rates? This study found 2 (yes, 2) "suspected" soft errors in 300 machines running for multiple months.

http://www2.ece.rochester.edu/~xinli/usenix07/

And not that Row Hammer should really be a concern for a workstation, but ECC does not necessarily protect against it. Some types of ram are vulnerable and some aren't. But ECC is not necessarily a factor.

http://www.thirdio.com/rowhammer.pdf

But that's not my argument (Even though I said ECC matters. maybe you're right).

My argument is that every time someone says "I can build the same thing for 1/3 the price", they are literally using completely different parts. They AREN'T building the same thing.

The idea of an Apple Tax is that apple arbitrarily marks up everything. But if you bought the same parts independently, you'd find the markup is very little. More like 10%.

Look at those $10k HP workstations. They are more comparable to Mac Pros.

But then the problem is just that Apple makes component choices with very poor value. A Threadripper CPU is not the same thing as the Xeon in the Mac Pro, but who cares? You can still build a way faster machine (with ECC) for half the price.

Apple techies hate to admit it. But in the end, it's the same reason people pay $2k for a plastic Louie Vuitton purse.

Except, at least with a fancy purse, you can carry it around and signal your wealth to others. My workstation is hidden away under my desk.

No I’d say a better analogy for what you’re trying to say:

I just want to buy a car for $20k that goes 70 miles an hour. But you’re only selling me cars for $50k cuz you insist on putting things like carbon fiber brakes and titanium alloys. I don’t need any of that. I can build something that perfectly meets my needs for way less. Because those enhancements are meaningless for what I’m trying to do (go 70 MPH)

Ya, or maybe even more precise:

You want to sell me a car for $50k that can go only 60 mph. But it has carbon fiber brakes and titanium alloys, requires premium gas, and has a fancy logo. Oh, and it has a flux capacitor which does nothing.

Or I could just buy a car that can go 100 mph for $20k.

Ya, it's not the same parts, but if the result is a higher performing system for way less money, what's the difference?

You can use different parts and end up with an equal or better outcome.

A lot of the Mac faithfuls point to ECC and Xeon to defend the price of the entry level iMac Pro.

Yet every single one of them I follow has used only MacBook Pros for the last 5 years or so which obviously don't have ECC so yeah I feel the importance of it is overstated.

This solution is fine for individual developers, or even small shops / startups, but it would never fly in a more corporate environment. Coincidentally, those environments are also the most likely to drop $5K+ on a development machine without batting an eye.

Not even close to true:

https://www.cultofmac.com/485914/imac-pro-pricetag/

Why would you need to buy a 1080 ti graphics card when imac pro has an amd vega. That makes the entire parts list suspect to me.

This list is a joke, that's not even close to a 1:1 comparison. It looks like they intentionally picked the most expensive options for everything. You don't need a $160 AIO water cooler, a $250 1000W power supply, or a $650(!?!?!) motherboard. The Vega in the iMac doesn't compare to a 1080 at all.

To anyone with an understanding of PC hardware it's clear this list was composed specifically to justify the iMac pro's price.

Here's a much more reasonable list: https://pcpartpicker.com/user/xanderstrike/saved/mNDWXL

This is for Logic and Final Cut users who want to have the best and don't know enough about computers to understand they're being ripped off.

Or they bill $2-300/hour so they don't really care. If their tooling is on the Apple side, they don't really have the option of going for a high end PC.

There's actually even more to it than just not caring. There is also the marketing side of it. Recording studios will sometimes make choices to buy certain gear because of the name and recognized brand. Apple computers are certainly the recognized brand for digital recording.

The software stack is very relevant, Apple shops would have to retrain to go PC, and some of the software they have is sub optimal.

My wife is a designer, she needs much more powerful computers than I do (as just a developer). She constantly runs out of memory, GPU performance is very relevant for Photoshop, Illustrator, and other tools she uses in her work. So spending $2500 or $3000 for a high end (but not pro) imac is a given, even though I would be perfectly happy with the $1900 one.

For her, the advantage in productivity is worth it. I can imagine there are many people who would see the imac pro as being worth it to them also and the price isn't going to bother them much because their time is already very valuable.

Reply on news.ycombinator.com