New Macs are out!

[pigoo3]

Nick, it's my understanding the new Retina iMac has a newer CPU in it than the previous so it might score even higher.

Yes...I was thinking about that too. Everymac doesn't have detailed info on the new one...so I was making my best theoretical guess.

- Nick

 

[Exodist]

Well if the new 27r iMac is a newer CPU. All things given if it follows the normal performance trend then at most its only 8% faster, all things being equal that is. Most people dont notice 8% anyway.

That said, next fall they could very well re-introduce a quad core Mini again. Part of the reason they may have elected to keep this one at dual core is so that it wouldnt take away from the 27" retina iMac sales.. But then we never really know what Apple is up to or could change in 6 months..

 

[pigoo3]

Well if the new 27r iMac is a newer CPU. All things given if it follows the normal performance trend then at most its only 8% faster, all things being equal that is. Most people dont notice 8% anyway.

Yes this may be true...and probably a good estimate. But the main point trying to be made in the last 3-4 posts was that the high end 2013 27" iMac...and now the 2014 27" retina iMac...both will have slightly higher cpu scores than the low-end quad-core new Mac Pro.

Sort of making them (wink...wink)..."poor-man's" Mac Pro's.

- Nick

 

[dtravis7]

For most users out there the iMac would be a better overall machine than the Mac Pro anyway. No messing with displays and other external stuff. The iMac is a very powerful computer with a great display which is even better with the new Retina display on that model! I can not wait for the reviews.

Really the top end iMac has been faster in CPU performance for quite a while than the Mac Pro. It's not a lot but if you get the best iMac it's a bit faster.

 

[pigoo3]

Really the top end iMac has been faster in CPU performance for quite a while than the Mac Pro. It's not a lot but if you get the best iMac it's a bit faster.

I watched a two part video by a guy from Germany on You Tube a couple weeks ago...where he reviewed the 2013 high-end 27" iMac vs. the quad-core 2014 Mac Pro.

The gist of it was...the 2013 27" iMac was faster (cpu) and I think comparable (gpu) to the 2014 quad-core Mac Pro. He was using the latest version of Cinebench.

His one uncertainly was (because the 2014 Mac Pro didn't blow away the iMac on the graphics test)...he wasn't 100% sure if Cinebench was actually written to test the Mac Pros dual video card setup properly.

He was on the 12th day of the 14th day return period...and he was trying to figure out if he should return the Mac Pro...get the iMac...and save a lot of money (around $1000).

His conclusion. Even though the benchmarks tests kinda were even (or slightly in favor of the iMac)...he decided that the Mac Pro had more future potential...due to all of the ports...and the ability to run multiple 4k or 5k large external monitors.

So I think that he decided to keep the Mac Pro.

- Nick

 

[chas_m]

Using one benchmark that's not tuned to specifically test the machine class you're using is frankly idiotic.

CPU is not the only factor, and testing the GPU with a test or programs not designed to measure workstation-class integration will never give you a real world approximation of performance.

From its ability to handle a great deal more RAM to programs optimized for that many cores or that can handle dual-GPU integration, the Mac Pro is simply not in the same class as the iMac, even now. Trying to judge it using consumer-grade tools is like trying to evaluate a supersonic aircraft using hang glider benchmark testing.

Just to prevent any misunderstanding, I'm criticizing the youtube guy from Germany. Ars Technica, he ain't.

 

[dtravis7]

I was waiting for Chas to respond and he did not let me down. The sites I go to use more than "ONE TEST" and know what they are doing. That is all I will say. Enjoy your Mac Pro!!!!

 

[Exodist]

The problem with general benchmarks is that they are normally geared for mains stream gaming hardware. The GPUs in the mac pro are workstation class and while they may suck playing world of warcraft even on low settings. They are made for CAD and other related software. I have seen $800 dual GPU nVidia geforce graphics cards run software like Solidworks and it run like complete crap. But take a 6 year old nVidia Quadro with half the related specs including RAM and it make Solidworks scream.

I have put a lot of thought into this and its likely I will just order 16GB of RAM from OWC and be done with upgrading until the next refresh of the iMac. Get my moneys worth from this system. Right now the most high end software I run is Final Cut Pro X and it runs fairly smooth most all the time. I think it will run smother when I put my FCP cache folder over on another drive. But stil even with 8GB of RAM, it runs great and only takes about 3mins to export to a 1080p MP4, 30min file.. So really no issues, just want faster GPU for play time.. LOL

 

[dtravis7]

Exodist, agree completely.

For most people that run single tasking applications, the IMac is as fast or faster. Found a really good review and tests that I will post later.

Where the Mac Pro excels is with Apps that run more than one task and Apps that use all the CPU Cores like say in a 6 Core Mac Pro.

Also the ATI Fire cards blow down the Imac with certain applications like Exodist just mentioned. For Gaming and everyday usage though in many tasks the top iMac is faster and a more sensible purchase. If you do heavy CAD, Final Cut Pro X and top end audio applications, get the Mac Pro.

https://www.youtube.com/watch?v=50oZLFbI7jQ

This video shows a lot of different tests and shows where the Pro excels and where the iMac excels.

 

[pigoo3]

Using one benchmark that's not tuned to specifically test the machine class you're using is frankly idiotic.

CPU is not the only factor, and testing the GPU with a test or programs not designed to measure workstation-class integration will never give you a real world approximation of performance.

Just curious. If you were in the same situation as this guy in Germany (deciding between the high end iMac or the entry level 2014 Mac Pro)…how would you have tested both computers performance to help decide which computer to keep (a data driven decision…not emotion)?

Thanks,

- Nick

 

[Exodist]

Just curious. If you were in the same situation as this guy in Germany (deciding between the high end iMac or the entry level 2014 Mac Pro)…how would you have tested both computers performance to help decide which computer to keep (a data driven decision…not emotion)?

Thanks,

- Nick

Well see, there is more to the Mac Pro then just performance. The whole system is based around Xeon CPUs and ECC RAM. The system is designed be reliable and prevent data corruption at the hardware level. This is generally these days server/enterprise grade hardware were data integrity ranks higher in priority then speed.

So if its absolutely important that your system never ever crash, this is the type of hardware you should be looking at. But if playing WoW or kicking back and listening to iTunes is your interested in, then this system is not for you. iMac would be better.

 

[pigoo3]

Well see, there is more to the Mac Pro then just performance. The whole system is based around Xeon CPUs and ECC RAM. The system is designed be reliable and prevent data corruption at the hardware level. This is generally these days server/enterprise grade hardware were data integrity ranks higher in priority then speed.

So if its absolutely important that your system never ever crash, this is the type of hardware you should be looking at. But if playing WoW or kicking back and listening to iTunes is your interested in, then this system is not for you. iMac would be better.

This really isn't an answer to my question. The question was…what sort of data driven method would someone use to decide whether to keep a high-end iMac or new Mac Pro.

Reliability is great…but your response doesn't answer the performance comparison question. When someone needs to crank out some projects with "need it yesterday" deadlines…a person wants as much performance as they can get (within the budget constraints that they have).

Let me put it another way. If someone was living in a cave (and knew nothing about iMac's & Mac Pro's)…they were given these two computers…and were told to decide which one to keep (keep the one with greater performance)…how would they go about deciding?

What data driven method would/could they use?

- Nick

 

[Exodist]

This really isn't an answer to my question. The question was…what sort of data driven method would someone use to decide whether to keep a high-end iMac or new Mac Pro.

Reliability is great…but your response doesn't answer the performance comparison question. When someone needs to crank out some projects with "need it yesterday" deadlines…a person wants as much performance as they can get (within the budget constraints that they have).

Let me put it another way. If someone was living in a cave (and knew nothing about iMac's & Mac Pro's)…they were given these two computers…and were told to decide which one to keep (keep the one with greater performance)…how would they go about deciding?

What data driven method would/could they use?

- Nick

Well Nick you answered your own question, if performance is paramount and your own a budget. The answer is of course is a iMac based on budget alone.

A better question would be if budget was not a concern and the person just wanted the fastest Apple offered. How would a person know which to get. And thats a really great question. Perhaps this is something Apple should be conveying to their consumer base. I know that the iMac is what most consider multi-media class system and the Mac Pro is workstation class system. But most will not. Matter fact the word "Pro" has been thrown around to liberal with the systems IMHO. Look at the Macbook Pro, its nothing like the Mac Pro, matter of fact its more like the iMac and higher end Minis.

When it all comes down to it, Apple needs to step up and clearly define what they consider "Pro" level gear instead of just flip flopping around with a overused marketing term.

As far as what can a person use (like a benchmark) to determine which is best for them. You cant! Its like comparing Apples to Oranges, sure both are fruit but they are completely different. One makes better pie, then other makes better juice IMHO..

 

[chas_m]

Just curious. If you were in the same situation as this guy in Germany (deciding between the high end iMac or the entry level 2014 Mac Pro)…how would you have tested both computers performance to help decide which computer to keep (a data driven decision…not emotion)?



Thanks,



- Nick

It's a simple application of common sense. What is the Mac Pro designed to do?

It's designed to work with very expensive, high-end software -- the kind of stuff that costs as much or more than the computer itself! High RAM requirements, full multi-core utilization, scientific precision, workstation 3D ...

An iMac,even this deluxe model, is aimed as a general-purpose computer. With the one difference being that the new one is pretty clearly designed, aimed and priced for the "creative pro" market.

So the question is really not "what are the benchmarks," the question to ask oneself is "what am I planning on doing with this machine?"

Both will run Photoshop, for example, but one is likely to have a total package that makes it the more obvious choice. Likewise, if I'm wanting to run mostly SPSS or advanced climate simulations or run renders of complex 3D, it's pretty obvious which one would do a better job.

 

[pigoo3]

It's a simple application of common sense. What is the Mac Pro designed to do?

Sorry…but this is not a data driven answer. A perspective buyer of a new computer needs data to decide what model new computer to purchase. Not a bunch of wishy-washy techno-babble.

And how the heck does the average computer purchaser apply what you call "common sense" to a computer purchase?? How fast, how slow, how large, or how small is "common sense"??

What is needed are test scores (large, small) or test times (how long or short) it takes to do something. Then the various computer models someone is interested in can be compared & contrasted. This is the way computer performance has been done for years & years & years…and I really don't understand why someone who is very computer knowledgeable would be debating this in 2014.

Sure benchmark testing apps may not be perfect…and may not accurately reflect what an individual computer user does on a daily basis. But then again…we can't have 10,000 different benchmark applications specifically written for how 10,000 different users use their computers.

And if benchmarking applications are so bad…why do professional reviewer's who write review articles for major magazines and websites use benchmarking application scores for performance comparisons??

So in the absence of something better…benchmark testing software is the best we can do as a way to fairly compare computer models BEFORE PURCHASING.

So again I ask…do you have a better method for comparing the performance of various computer models BEFORE purchase…so a customer can purchase the best computer for their needs??

"Common Sense"…is really not something a computer purchaser can use to compare the performance of computer models.

- Nick

 

[dtravis7]

Nick Agreed 100%. The one video I put up does what you asked for. Seen other tests like that and results were the same.

Like I said earlier, the Pro excels with Multi Threaded and the iMac with Single Threading.

And also with those pro video cards they excel at certain uses like Cad and other Pro applications.

 

[chas_m]

Sorry…but this is not a data driven answer.

In fact, it is. But apparently not the data you were expecting.

All computer purchasing -- unless you're a moron -- is driven by finding the best machine for your needs. There's absolutely no "wishy-washy techno babble" in my post, and I look forward to your apology.

And how the heck does the average computer purchaser apply what you call "common sense" to a computer purchase??

By asking themselves "what is it that I want to do with this computer?" That should ALWAYS be the starting point.

What is needed are test scores (large, small) or test times (how long or short) it takes to do something.

NOPE.

Really, Nick -- you of all people -- a slave to a spec sheet? I assume you must own a PC and an Android phone then -- after all, they have better specs!! QUAD HD MAN!!

CPU scores in particular are nearly worthless in isolation, because different sorts of machines are optimized (oh dear, there's that techno-babble again!) for different purposes. IBM's Deep Blue CAN do your taxes, but really is that the right machine for doing taxes?? According to you, it is -- because wow look at those CPU scores!!

Do you buy a car based solely on its 0-60 speed? Solely on its gas mileage? Soley on cost?

Or ... maybe, just maybe ... are there several other factors beyond a given spec that go into a wise decision?

 

[dtravis7]

First I do not like you using Idiotic and Moron when addressing people here. It really sounds like you are calling us that.

Second, did you watch the video I put up? Actual applications doing real work yet you ignored it. I would have loved to debate you back in the days of the Power PC when with a G5 even conversion of Videos I was being PAID to get done took sometimes 1 hour where the 1.66Ghz Intel Mini with the SAME APPLICATIONS took under 20 min. When I have work to GET DONE, Speed is very important. over 1 hour Vs 20 min? Add that up with maybe 10 videos I had to have done and done yesterday and tell me which machine you would use?

I had Apple Worshipers tell me the G5 was better, Better at what? Wasting my time? Getting my clients mad as they had deadlines and needed the work?

Don't be a blind Apple can do no wrong please!

 

[dtravis7]

So Final Cut is a CPU BenchmarK? Premier is a CPU Benchmark? Really?

 

[pigoo3]

In fact, it is. But apparently not the data you were expecting.

All computer purchasing -- unless you're a moron -- is driven by finding the best machine for your needs.

Again...this is not an answer that the average computer purchaser will find helpful if they are in the market to purchase a new computer.

Imagine an average computer user shopping for a new computer. And they come up to you & ask..."What would be the best way for me to compare the performance of computer A, computer B, and computer C?"

And your answer to this person is a combination of what you said in your last two posts:

1. It's a simple application of "common sense".
2. All computer purchasing -- unless you're a moron -- is driven by finding the best machine for your needs.

How do these two "pearls of wisdom" help someone purchase a computer...or assess the performance of 2-3 computer models they are considering??

There's absolutely no "wishy-washy techno babble" in my post, and I look forward to your apology.

Again...you are using your measuring stick to determine what is & what isn't "techno-babble". The techno-babble I'm referring is this quote (below):

It's designed to work with very expensive, high-end software -- the kind of stuff that costs as much or more than the computer itself! High RAM requirements, full multi-core utilization, scientific precision, workstation 3D ...

This statement (quote above) made to the average computer purchaser will just result in their eyes glazing over...and their head spinning. To the average computer purchaser...this is "wishy-washy techno-babble"!

To you maybe this isn't techno-babble...and this is why you are not able to put yourself in the "shoes" of an average computer purchaser. You're using your own measuring stick...and assuming that if you understand it...then everyone should understand it. And this just isn't so. Thus you are not able to help them.

By asking themselves "what is it that I want to do with this computer?" That should ALWAYS be the starting point.

I agree...this can be something a person asks themselves as a STARTING POINT! Now we need some data (benchmark tests) to answer many of the remaining performance related questions.

Really, Nick -- you of all people -- a slave to a spec sheet? I assume you must own a PC and an Android phone then -- after all, they have better specs!! QUAD HD MAN!!

CPU scores in particular are nearly worthless in isolation, because different sorts of machines are optimized (oh dear, there's that techno-babble again!) for different purposes. IBM's Deep Blue CAN do your taxes, but really is that the right machine for doing taxes?? According to you, it is -- because wow look at those CPU scores!!

Do you buy a car based solely on its 0-60 speed? Solely on its gas mileage? Soley on cost?

Or ... maybe, just maybe ... are there several other factors beyond a given spec that go into a wise decision?

Again you're losing yourself by focusing on the wrong things!

The situation is...Joe average computer consumer wants to purchase a new computer which costs a lot of money. Joe average computer user wants to get the most computing performance for his money. If computer benchmarking aplications are not useful (according to you)...what objective way can he use to make a decision to assess the performance of the computer models he is interested in?

There's nothing in your last two posts that would be of much help to Joe! You haven't given him any real objective data that says computer A is better than computer B or computer C (from a performance perspective)...that's a better method than using computer benchmarking applications.

- Nick