Sabtu, 27 Agustus 2016

Silicon Week: Alienware: 'We didn't build the Graphics Amplifier to be a one-hit wonder'

Silicon Week: Alienware: 'We didn't build the Graphics Amplifier to be a one-hit wonder'

Alienware has now been going for more than 20 years, and it's been a decade since the company partnered with Dell to work together on increasingly ambitious gaming machines. The Miami, US-based company makes aspirational products loaded with power, built with otherworldly designs and carrying premium price tags to match.

According to its co-founder and General Manager Frank Azor, Alienware has fostered an ambition innovation and had a drive to make the best gaming PCs in the business from the very start, even if its core team didn't quite know what they were doing in the early years.

We caught up with Azor to dig deep into the company's history while finding out about newer products - from the Alpha R2 Steam machine to its Graphics Amplifier, Aurora R5 gaming desktop and VR backpack.

TechRadar: Alienware has been going for more than 20 years now. It was the first company to launch an 11-inch gaming laptop, build a Steam Machine and create an external GPU enclosure. Which "first" are you most proud of?

Frank Azor: I think it has to be the launch of the Predator chassis, the desktop case we launched that really defined the look of Alienware back in 2002. I would say that's the proudest and most significant moment in our history. We launched that at a time when pretty much every computer except for maybe Apple ones were beige, boring boxes - or just a black box with very little industrial design attributes.

Alienware

Design wasn't really a factor in PCs prior to us introducing that product. You look now and design is a critical factor in any premium product out there. The Predator was imitated by so many companies for so long, and that's hugely flattering.

Did you know what you were doing right from the start?

Behind the scenes, a lot of people don't know that we didn't know what the hell we were doing - the Predator was really the first product we developed ourselves from the ground up. We had to learn a lot about product development and engineering. We also had to hire the right people and really take a completely different approach to how we were introducing products and engineering them because prior to that we were mostly doing a really good job as hobbyists - buying great components off the shelf, putting them together, testing and validating them.

How do you approach a new design? Do you start with a problem and design around it, or come up with a cool design and see how it could solve a problem?

There isn't one way that we do it. We're very innovation-focused and allow ideas to come from anywhere within an organisation. We have forums and meetings and ways of engagement that help foster the sharing of those ideas. They may occur at our annual strategy and gaming session that we do at Alienware's Miami HQ, or they may come from an industrial design meeting or from our customers.

We run with those ideas and deliver solutions like the external Graphics Amplifier, or things like G-Sync panels. There isn't a central innovation organization where all the ideas come from - I think that would be very limiting to the creativity process and the potential of the entire organization.

Some people think that Steam Machines have missed the boat. Why does Alienware think that they have a bright future?

My perspective is that Steam Machines have been a phenomenal success, but it depends how what you measure that by. If we remember the PC ecosystem prior to Steam Machines, we had a Windows 8 environment. That was an operating system Microsoft admitted wasn't very focused on the PC gamer at the time. As a result, the gaming community was a little upset by that. Steam Machines then came along and aimed to build and operating system for the PC gamer. After that effort came Windows 10, with DirectX 12. Microsoft really woke and knew it needed to re-engage with PC gamers or really risk losing them.

The Steam Machine controller

As such, I would give some credit to the Steam Machine initiative for helping Microsoft realize what they had. If you look at the actual use model of Steam Machines, there's a lot of benefit to the platform. If somebody isn't interested in dealing with Windows updates or using a keyboard and mouse - somebody who might just want an appliance on their TV to play thousands of games out there - then a Steam Machine is a solution.

Do you think there's potential for Steam Machines to ever sell in the millions of units?

They may never sell in the millions of units, quite honestly. But the fact is that there's a solution and there's been a huge partnership between hardware companies and Valve to go and build these solutions. I think this is the first time we've had an alternative platform that's viable, works, is reliable and is shipping.

Alienware Alpha R2

The industry, depending on how things evolve over time, may gravitate towards that in the future. If it doesn't then I would say that's also a win, because it means that traditional Windows PCs are doing so much right that there's really no need to move over to Steam Machines. I think that's great for the entire industry. On the other hand, if things don't continue to go well for Windows PCs, then there's another option for the entire industry and that's a great win too.

From the Alpha R2 and the Graphics Amplifier to the Aurora R5 or the X51, Alienware's recent products have a clear upgrade path. How important is that in the age of VR, where it's inevitable that gamers will need increasingly powerful GPUs as headsets evolve?

Upgradability has always been a key factor for us. The reality is that it's probably a better business decision to not make a product upgradeable and some companies make those decisions. When we started the company in the early days we always had the mantra of building things the way we would want them built for ourselves.

Aurora

Take our notebooks, for example. For a long time everything in them was upgradeable. Over time certain things have unfortunately moved beyond our control and changed in the industry. For example, silicon has moved from being socketed to being soldered on as the only option, which has limited our customers' upgrade path in our notebooks. It's not because we want that - we've pushed our vendors to do things the old way, but some decisions are beyond our control.

What impact do you think the Nvidia's new GTX 10-series notebooks will have on external GPU enclosures? Do you think people will gravitate toward laptops with that power there, rather than a weaker solution with an external booster?

Well, we didn't build the external graphics amplifier to be a one-hit wonder, or a solution to a period of six months or a year. It's a commitment we've made for a very long time. We were the first out there with such a solution and have developed a proprietary interface around it because it has significant bandwidth and cost advantages for the customer.

Alienware Graphics

There's confusion around external GPU enclosures too - people think they can plug them into any USB-C notebook and they'll have external graphics support, but that's not the case at all. Our investment in external graphics has been a long term commitment of around eight years or so, going by our patents. This is a long-term vision for us and something we've looked into significantly.

It's great that notebook performance has mostly caught up with desktop graphics. Doing VR in all our notebooks and the value you get for dollars per performance is all phenomenal, but when the next generation of notebooks comes along in two or three years for now, you've got two options; buy a new notebook for thousands of dollars, or a Graphics Amplifier for two-hundred bucks, along with an off the shelf graphics card. For less than half the money you just got a jolt in performance for 4K or 8K gaming, or VR, whatever it is.

What are the performance advantages of running, say, a GTX 1070 in an external GPU enclosure versus having it in a notebook?

Even though you can buy a gaming notebook with a 1080 in it, it's a big GPU. The maximum performance that you're going to get out of that mobile GPU is very close to the standard clock performance of a desktop 1080, but the thermal headroom and overclocking abilities of the desktop 1080 is considerably higher than that of the mobile version of the same card.

GTX 1080

Notebooks will always be playing catch-up compared to what you can do in desktops, and that's because you have more room to work with inside ad desktop card - the thermal solution is bigger and you have more flexibility - and you can cool it differently. So I'm not worried about any short-term implications - demand for Graphics Amplifiers may slow down a bit now that the 1080 is out, but when 11-series cards come out, or whatever Nvidia calls them, the cycle will repeat itself. The great thing is that we have a solution and customers have an option - and they've never had that before.

VR backpack

Alienware's VR backpack is great-looking device. When will it be available to buy?

We showed it off at E3 and asked folks in the Alienware arena if they'd be interested in one. We invested in making the prototype, and if people want it commercialized then we're still listening. It's a category that we're really interested in developing in the future - we just need some support from our customers and the parties that are interested. Then we'll know whether it's the right decision to make or not.

This article is part of TechRadar's Silicon Week. The world inside of our machines is changing more rapidly than ever, so we're looking to explore everything CPUs, GPUs and all other forms of the most precious metal in computing.


August 27, 2016 at 03:00PM
Kane Fulton

Tidak ada komentar:

Posting Komentar