Hacker News new | ask | show | jobs
by greenspot 3448 days ago
If you are a designer, this is absolutely the best thing you can do to sky-rocket your market value over night:

Create a concept design from a popular product and put it on a slick landing page. It shows that you, as a designer, are proactive and think beyond designing standard stuff (like webpages or mobile apps).

Moreover, you are not limited by any client restrictions[1] which hurt your work (and portfolio), you learn 3D modelling if you haven't yet (it's not hard just time consuming), if you are lucky with social news sites you get so much free promo and finally, it's the eye-catcher on any CV.

[1] A classic and recommended post if we talk about clients restricting designers: http://theoatmeal.com/comics/design_hell

12 comments

> Moreover, you are not limited by any client restrictions

But you must still limit yourself to sanity and physics. Too often, designers don't have a clue what's actually going on behind the scenes, and embarrass themselves.

This example puts 16 TB3 ports on a computer, because "consumers want it to be very expandable" and "2 columns of 8 looks pretty". It also adds two graphics cards but buries the connectors into the floor rather than making them externally accessible.

Concept cars by designers may have faults such as zero visibility from the drivers seat, have aggressively high front bumpers and low hood lines that look designed to kill pedestrians, utterly lack necessary things like exhaust pipes, crumple zones and spare tires, or have absurd specifications ("500 miles from the 2 cubic foot battery pack!" "600 HP V12 under the rear seat!")

Your target (other marketing departments) may not care. But it's also very possible that they encounter these limits as a part of their daily work, and will care, judging you for your lack of domain knowledge. Definitely create some concept work for your portfolio. But don't stray too far from the realm of the possible.

Ugh - did anyone take a look at his portfolio? He designs for video games so of course the specs are exaggerated.

He did something fun and most of HN is railing on him. The same criticisms ought to be said here - if you're going to criticize, first take a look at the whole picture and don't jump to conclusions.

I agree that the ad hominems are unfair, but bad design is bad design regardless of authorship.
But part of me thinks he's an armchair product designer. Anyone can create a cool looking mockup. But what about the actual thing?
Anyone, huh?
Even as a non-designer consumer it sucks when you can look at a concept and immediately see that it wouldn't work in real life.
It's weird that they presented the cards as standard PCE-E cards when the existing Mac Pro uses custom cards that don't have their own cooling manifold, but I don't think their intention was that there'd be connectors on the bottom you'd use.

Just like the current Mac Pro I'm sure they expect you'd use Thunderbolt or HDMI for display out. Those connectors are the same as on the existing Mac Pro, except this one has TB3 as well.

I hadn't even noticed the floor facing video card connectors... I imagine this could be addressed by having mini display port cables that connect to external ports on the back. I agree that VGA and DVI would likely not fit. From what I recall most modern video cards have multiple output types though.

I know L connectors exist for hdmi too: https://www.monoprice.com/product?p_id=3733 and the category in general: https://www.monoprice.com/category?c_id=104&cp_id=10419&cs_i...

heat is exsausted out the connector side on modern GPUs, placing that facing downward is an excellent recipie for a cooked GPU.
Just reverse them (so the GPUs are venting up) then have right-angle dongles (which are very Apple now) under a magic magnetic lid compartment. Or, recess them enough (increase the case height) enough that you can plug in normal DP / HDMI cables.

I like it, but it's not proprietary enough for Apple to make. They didn't come out with a single upgrade GPU for the Mac Pro dustbin design, so I doubt they care much about selling modularity or upgradability these days. Their message is clear: "upgrade" means new device, which means all the products they sell are disposable appliances.

IMO, they could open up MacOS to work on some variations of hardware and sell it for $300-$500. They would still make the nicest hardware, but since desktops and laptops obviously aren't their main focus anymore, they could still offer power users good tools to build all these amazing mobile apps on without limiting them to their current slim pickings. Hell, even just make XCode cross-platform. Even just let it run on X, and use the Windows Linux subsystem to support it.

Using two standard full-length cards is just lazy design (even if he made it a point to use standard components). The GPUs should be on boards like they are on the current MacPro and attached to the monumental heatsink. This add two fans to a design that absolutely doesn't need them. The SATA disks also seem misplaced - it should extend PCI-E storage or, even better, DIMM flash modules: big Xeons have 4 memory channels per socket, enough for letting a couple of those go to storage. Having one single bus for everything makes the machine much neater.
> The GPUs should be on boards like they are on the current MacPro and attached to the monumental heatsink

This is the exact problem with the trashcan, it's a computer made for GPU computation (one of the trashcans GPUs isn't even hooked up to the display it's purely for computation) yet you are stuck with proprietary versions of GPUs that run the wrong sort of code.

Almost the entire GPU computing community works on CUDA, the trashcan cards don't run CUDA and because they're proprietary they can't be upgraded.

The right solutions are either this design or one where there is a heatsink+liquid cooling bracket system where you can attach a GPU after taking the stock cooler off to. Which is also very common thing to do for people in GPU computing. The latter invalidates your warranty so this design actually makes sense.

Just like the designers of the current Mac Pro, you don't get it.

This design (even though it doesn't work in reality) is about creating something that is functionally like the earlier Mac Pros. That means commodity off-the-shelf hardware can be used, not just specialized or outlandish components that are going to be very expensive, if anybody bothered to produce them at all. The current Mac Pro still ships with the outdated GPUs it originally was introduced with, with no upgrade path foreseeable.

Case in point: their 'concept gun' ignores the fact that your trigger hand is attached to an arm behind it.
Heh, I work in a design studio and would temper your statement. Yes, having something like this in your portfolio certainly can't hurt - and having it frontpage HN is most certainly a good thing, exposure wise.

But when we interview candidates and do portfolio reviews, a project like this would be one of the things I'd be the least interested in. It does showcase the designer's creativity and skills at using 3D software, but it doesn't give you any sense of how they can work in a team, with clients, with real specs/requirements to follow, etc. It's kind of like the kind of designs that architecture students do in school: sure, the end designs might look very cool, but they don't represent capacity to work with the real world architectural process in any way.

One of the skills I value the most in a designer is the ability to really think through the various sub-aspects of a problem under different angles, come up with many alternatives, etc. A concept design like the one posted does not show this at all.

If you are a designer, your portfolio should only have one conceptual piece of this sort at the very most, and it shouldn't be its main focus. A designer's main skill should be providing solutions or insights into real world problems, and this doesn't solve any meaningful problem other than "make something that looks cool and will never exist".

It's pretty clear you and he are at opposite ends of the design spectrum. He's an experienced conceptual designer perfect for video games, i.e. Art Center. You're more academic and instructional, i.e. MIT.

Both are legit.

PS - it's pretty clear he's just having fun with the Mac specs. Only kneejerk HNers are taking it as a serious product pitch. He's designed beautiful guns, vehicles... why not a Mac that makes Ives look seriously plugged-up?

The thread is discussing the value of this type of conceptual presentation to a design portfolio. 'Conceptual designer' by your definition would basically mean 3D visual artist. That's not quite 'design' as it's being discussed here, though I wouldn't call it academic or instructional either. For design work, the final product rarely demonstrates the skills needed to be a good designer in real business settings. Hiring managers want to see the process that led to the final product.
You might be right. But if you're a consumer, these kind of concepts do nothing but frustrate because it:

1) sets totally unrealistic expectations

2) is total fantasy

And lesser informed people often think these are real if they see them. I recalling having had multiple people tell me about the "iPhone holographic keyboard" from some silly concept porn a few years ago...

> is total fantasy

Oh come on, can't we imagine things that don't exist for fun?

Yes, but suspension of disbelief is hard (and even harder for us here). The objects you design need to be able to do their jobs or have convincing explanations for why they look like what they do.

Moebius got away with a spaceship shaped like Dumbo from Disneyland's Dumbo the Flying Elephant ride because the technology that surrounded it was so advanced that a spaceship shaped like that was no big deal. Seemed perfectly doable.

Sure, that's why it's called fantasy.
No.
Where's that sarcasmark when you need it!
I might argue that you're right on most counts, but there's one project type that'd be a step above that: creating a side project that stands entirely on its own.

It has the effect, like you had said, of being not limited by clients and learning new skills. But on top of that, you'd show yourself as an independent thinker and started closer to zero than standing on the shoulders of other companies (in this case, Apple) and making spec work.

Again, not to say it's wrong, because I think you're very correct in thinking this is great for designers, but I do think this could be one step higher.

(And, for the sake of argument, I'm completely ignoring what neither of us stated: that probably more than half the battle as a professional in this industry is navigating politics, which the skills for obviously cannot be conveyed through a spec or side project.)

Reminded me of what Dustin Curtis did for American Airlines. http://www.flipthefunnelnow.com/dustin-curtis-%E2%80%93-dear...
I liked AA employee's response: ...the thing I most wanted to get across—simply doing a home page redesign is a piece of cake. You want a redesign? I’ve got six of them in my archives. It only takes a few hours to put together a really good-looking one, as you demonstrated in your post. But doing the design isn’t the hard part...
The guy who did minimallyminimal.com did this several times, and it (probably) got him hired at Microsoft. It was interesting to see the progression of quality and scope of his work as I was following the blog since quite a long time before it took off.
He was also a top student at one of the best industrial design schools in the world, and had some stellar internship experiences.
Reminds me of the designer who redesigned Microsoft's branding [0], and got a job there [1].

[0] https://news.ycombinator.com/item?id=4195208 [1] https://news.ycombinator.com/item?id=5085915

Trust me. This 3d modeling, material design, lighting, etc. is NOT trivial, lol. AND its time consuming!
Don't forget to find hosting that doesn't buckle when you get front-paged.
that's called a static site :-P
It really depends on the server. You can crush even something as resilient as Nginx serving static files if you try hard enough.
Sure, but it's orders of magnitude faster than dynamic sites. Additionally, stick CloudFlare in front of it and it's 100% fine for 99% of sites :)
Just make sure to count your ports properly. :)
How much time would it take for a professional designer to put something like this? What tools would you use?

I'm not a designer and just curious.

Depends on if you want it to just look good or if you want it to be technically feasible from electrical, thermal, and manufacturing standpoints. Taking all of that into consideration would easily take 10-100x more time than just a novel visual prototype. Even look at Razer's triple-screen laptop from CES; they had one which worked, and one which looked pretty. Getting that last 10% of fit, finish, and feasibility takes a long time too.
If I used Fusion 360 (free for personal use) to design this, it would probably take me tens if not a couple hundred hours. Maybe you can get pre-made parts, like the memory chips, heat sinks and stuff.

Like another commenter said, fairly easy to do, but it takes serious time and creativity.

Why do people say this stuff is easy? It takes time and a lot of practice to understand the nuances of materials, lighting, how light interacts with various materials, cinematography, and not to mention the ins and outs of 3d modeling software.
What do you think the equivalent move is for front end developers?