News for hackers. Stuff that matters.
Re:Coke computer vs. Pepsi computer by raganwald on Tuesday November 15, 2011 @10:28AM (#3238501)
The essential point about specs is bang on, but comparing computer purchasing to soft drinks over-trivializes it. Soft drinks have nearly zero feature differentiation.

Lots of consumer goods have meaningful features. One dishwasher might be quiet. Another might use very little energy. A third may heat its own water, sanitizing the dishes. You may want a model that can be fitted with a panel to match the cabinetry.

What dishwasher manufacturers have moved away from is specifications that do not strongly correlate with features. How many horsepower does a dishwasher motor have? How many chips are in its circuitry? What does it weigh? These things are rarely front and centre, because most buyers can't draw a connection between them and a benefit they will enjoy.

The trend is not towards brands being the only thing that matter, but rather towards features that probide benefits. RAM and Megahertz are specifications. Being powerful enough to do speech recognition in real time is a feature. Siri is a benefit.

Re:Coke computer vs. Pepsi computer by libria on Tuesday November 15, 2011 @11:59AM (#3238940)
Some specs matter; the problem is that they don't present their strengths clearly or they dilute it with meaningless ones. I could send a soccer mom to a store and tell her to prefer an SSD over 4200 RPM, i7 over Atom, 4GB RAM over 512MB. She doesn't know what these are, but she does know which one is better, the same way she'd know that a V6 is more jumpy than a 4-cylinder.

(edit, had RAM specs swapped)

Re:Coke computer vs. Pepsi computer by ugh on Tuesday November 15, 2011 @12:51PM (#3239213)
Specs are an awful proxy.

Specs do matter insofar as they impact user experience. That relation, however, is not linear. Diminishing returns are very common and there are also discontinuities.

Also, there are tradeoffs to consider: noise, temperature, weight, size, speed, battery, price, software …

You can’t have it all. That’s just not an option.

Specs are an implementation detail. They matter, sure, but the average consumer shouldn’t try and judge products based on them. That’s destined to fail. They should look at user experience and tradeoffs.

Companies and reviewers should try to translate the specs that were used to implement tradeoffs into user experience. Who is this device good for? What tradeoffs does it try to pick? Does it do a good job at that?

That’s what matters.

Re:Coke computer vs. Pepsi computer by jiggy2011 on Tuesday November 15, 2011 @01:08PM (#3239305)
The problem with doing that is it would require very comprehensive reviews or for customers to very thoroughly try out all products before purchase.

For example if a new product is launched on the market and it appears to be cheap but you know that it has a Fast Quad Core CPU and decent video hardware , you already know in a relatively small amount of text that it will be good for video editing , gaming and various other tasks.

If it only has a 300Mhz CPU you already know that it will not be good for any of these tasks and you can confidently avoid it.

As another example I remember years ago seeing computer adverts which would attempt to simplying their specs with bylines such as "Family Computer" , "Small business computer" , "Games Computer" etc. What if you want a PC for use by all the family for homework/internet etc but the oldest child happens to be an avid WoW/COD Player and Dad plans to use it to prepare accounts for his small retail business?

Re:Coke computer vs. Pepsi computer by ugh on Tuesday November 15, 2011 @01:11PM (#3239328)
It’s not so hard, really. There are a very limited number of use cases, a limited number of tradeoffs. A long list of specs is much worse than just writing “This machine is great if you need a lot of power for video editing and rendering and other creative tasks. It is, however, correspondingly noisy, hot and big, not a very good fit for mobile use or for anyone who needs a quiet computer.” That’s so much more helpful. Specs are code in comparison.

Specs confuse, specs frighten, specs suck. They force consumers to figure out the tradeoffs that stand behind them. That’s too much work.

Re:Coke computer vs. Pepsi computer by jiggy2011 on Tuesday November 15, 2011 @01:29PM (#3239414)
Not necessarily , the specs do confuse but that is mainly because the computer market as it is at the moment is confusing.

Just to pick on the example you used, let's say I want to write a novel and create a flyer for a local charity event. These are "creative tasks" how do I know if this computer will be able to do that, do I really need a $2000 Mac Pro for these?

It is big, but how big? Will it fit into the cupboard next to my desk?

Computers have a huge number of use cases that will continue to grow rather than decline, listing the suitability of a computer for each one of these would take far too much space in a brochure if it didn't computer manufacturers would have been marketing this way already.

(Edited for reply to next post)

I think my point is that it would be very difficult , almost impossible to write comprehensive reviews without any specifications to work from, especially brochure/advetising which needs to be very short.

Essentially a good spec is the normalization of a description of it's suitability for different use cases.

For example a computer which can run Angry Birds happily but not Call of Duty MW3 at all.. Would you describe this as a "poor games computer"?

If all I want to play is Angry Birds (and similar games) then it may be a fantastic games computer , if I want to run COD then it isn't a suitable games computer at all.. At least a spec listing allows me to look at the minimum requirements of the software I want to run and know whether it is suitable or not.

Video rigs would be overpowered for other work, but we only know that because we understand the requirements of video editing on a computers hardware. A layman will not necessarily know this but they will know that their computer meets the recommended requirements for Microsoft Word and Publisher but not for Final Cut Pro because it "doesn't contain enough ghz".

Re:Coke computer vs. Pepsi computer by ugh on Tuesday November 15, 2011 @01:37PM (#3239448)
I’m a terrible review writer, really. Don’t judge me.

When I say specs I’m mostly talking about internals. Size and weight do matter very much and those are also two things people actually understand†. CPUs, GPUs? Not so much.

Other than that you need only a better description. Video rigs are overpowered for other creative tasks, the review wouldn’t read as I wrote it.

—

† And now you can argue whether it’s better to actually write down numbers or to compare devices with aircraft carriers and football fields. I think both can be appropriate at times.

Re:Coke computer vs. Pepsi computer by sliverstorm on Tuesday November 15, 2011 @01:10PM (#3239320)
a V6 is more jumpy than a 4-cylinder

What?

Re:Coke computer vs. Pepsi computer by eykanal on Tuesday November 15, 2011 @10:56AM (#3238616)
Your point about over-trivializing is a good one. I had a hard time figuring out what sector to use as an example.
Re:Coke computer vs. Pepsi computer by ovi256 on Tuesday November 15, 2011 @11:11AM (#3238694)
Movies. Handtools. Paintings.

They may seem a disparate set, the connexion is subjectivity. Movies and paintings both speak in different ways to different people, and everyone knows they may like something their closest friend hated.

It's even more of a clear cut with handtools. If you don't get the right one to fit your task, any talk of quality/speed is useless. The best concrete drill can't replace a carpenter's hammer if what you need is to extract nails.