Comment

Unfortunately, Virgin Galactic’s trips are sub-orbital. :(

12 days ago on Bieber in space: popstar will embark on Virgin voyage 2 recommends

Comment

Alright, it seems something’s changed in OS X since the last time I tried the T221, because this time it Just Worked. (Last time was on 10.7.4, now I’m on 10.8.3.)

I got 24 Hz without even trying, using Retina DisplayMenu. (System Preferences couldn’t get it, though.)

Then, with SwitchResX, I got both 27.612 Hz, as well as the true maximum of 33.839 Hz (but that had strange shimmering artifacts, due to me pushing the Monoprice adapter too hard).

13 days ago on Asus' $4,000 4K monitor goes on sale this month, pushes Retina MacBook to its breaking point

Comment

whoosh

Seems Marcos needed a sarcasm tag on that post.

14 days ago on Google's lobbying budget is eighth largest in US, surpassing even Lockheed Martin 1 reply

Comment

Although, the 650M in a MBPR is overclocked significantly from the factory. A “real” 650M, which is what Intel is comparing to, is not 900 MHz with GDDR5 like the MBPR one, but is rather 835 MHz with DDR3, or 745 MHz with GDDR5.

A normal 660M is 835 MHz with GDDR5, for comparison.

In other words, a real 650M is 83% of the speed of the MBPR one.

14 days ago on Asus' $4,000 4K monitor goes on sale this month, pushes Retina MacBook to its breaking point 1 reply 3 recommends

Comment

Yeah, I need to pick up one of the Accell 3D adapters.

Just checked, and there’s a 233 MHz 24 Hz mode in my adapter, though, based on a posting I made a while back, so I would’ve tried that. I wasn’t running Boot Camp at the time, though.

I need to drop some things off at my storage unit tonight, so I’ll grab the T221, a cable, the LFH-60 to DL adapter, and the DP to DL adapter when I’m there.

14 days ago on Asus' $4,000 4K monitor goes on sale this month, pushes Retina MacBook to its breaking point

Comment

The Monoprice adapter is rated at a pixel clock of 268.5 MHz, which should do 3840×2400 @ 27.612 Hz. (Actually, maybe that’s the problem, I forget if I tried that mode specifically. I know every other dual-link mode I tried to shove in it was rejected, though…)

14 days ago on Asus' $4,000 4K monitor goes on sale this month, pushes Retina MacBook to its breaking point 1 reply

Comment

I wonder what the problem is – I’d guess that the monitor doesn’t have an EDID entry for a 3840×2160 mode that fits in the bandwidth limitations of the MBPR’s video outputs.

I’ve personally driven 3840×2400 off of single-link DVI off of a Thunderbolt port on the MBPR 15, so I know it can be done. (I wasn’t able to get dual-link going on the Monoprice dual-link adapter I was using, though, despite having the correct hardware to do dual-link on a T221. OS X was simply rejecting the mode that I had made in SwitchResX.) Mind you, I was doing it at 17.25 Hz or something like that, and you WILL see mouse pointer lag in that mode… it also took a few seconds for the monitor to gain proper horizontal sync.

14 days ago on Asus' $4,000 4K monitor goes on sale this month, pushes Retina MacBook to its breaking point 1 recommend

Comment

Never mind that if Iran’s government wants computer equipment, they can get it easily enough through enough layers of loopholes, even from US vendors. Sure, there may be some laws broken, but…

19 days ago on US removes sanctions on computer exports to Iran ahead of next month's elections

Comment

A “750M” SKU, with Apple getting the best bins of the chip and running it at 760M+ speeds (much like how they’re doing the “650M” in the current machine), could work, though.

The “650M” in the Retinas now is 384 CUDA cores, 900 MHz base clock. Official speed for a GTX 660M (same chip, same number of CUDA cores as the 650M) is 835 MHz.

Assuming the 760M could be binned to Apple’s 45 W TDP requirement, it’d be ~1.46x the speed of the current “650M”.

19 days ago on Nvidia's new GPUs promise to take your laptop gaming to 1080p and beyond 2 replies 4 recommends

Comment

I don’t think it’s ACTUALLY a wasted vote either.

28 days ago on Fox News reporter was labeled a 'co-conspirator' to justify DOJ email probe (update) 1 recommend

Comment

I voted for Jill Stein. (And I forget who I voted for before that, but it wasn’t Obama or McCain.)

/votingonlydoesnothingifyoudon’tthrowyourvoteawaybyvotingforthirdparties

28 days ago on Fox News reporter was labeled a 'co-conspirator' to justify DOJ email probe (update) 1 reply 1 recommend

Comment

Well, if you’re working text-centric, sure. But, em/rem is font-size-based, font-size is typically pt based (but can be any other HTML unit), pt is defined in terms of inches, and inches are defined in terms of px.

Ultimately, it all boils down to px.

29 days ago on Samsung beats Chromebook Pixel and Retina MacBook with new high-res laptop display

Comment

Well, NHK has proven that at 16", someone with 20/20 vision can actually get improvements to image quality all the way out to 1110 ppi.

29 days ago on Samsung beats Chromebook Pixel and Retina MacBook with new high-res laptop display 1 reply 1 recommend

Comment

In theory, HTML uses the px as the basic unit, and the inch is a defined unit of the px (that is, 1 inch is defined as 96 px, not 1 px is 1/96 inch.)

So, you simply change the px to pixel mapping to correspond with the actual display’s density, and let the OS/rendering engine do the rest of the magic.

Problem is, software doesn’t actually support this properly, and there’s no standard to request an image in certain pixel dimensions (Apple stuff, you can use some JavaScript to grab @2x images), although you can always put full res up, and then use HTML height and width tags to scale it down (to the appropriate actual size, and the client gets all of the detail available) client-side if you don’t care about the bandwidth usage…

29 days ago on Samsung beats Chromebook Pixel and Retina MacBook with new high-res laptop display 1 reply

Comment

I’ve fed a 3840×2400 display off my MBPR, worked fine (for some values of fine – the hackish setup that I’m using means that it was a bit buggy, but when I wasn’t triggering those bugs (display going to sleep, or the first few seconds after plugging it in, and both of those I’m going to blame on the Monoprice mDP→“DL”-DVI adapter (couldn’t even get dual link working)), it worked fine).

29 days ago on Samsung beats Chromebook Pixel and Retina MacBook with new high-res laptop display 1 recommend

Comment

The problem with Thunderbolt is that its video side is merely DisplayPort.

Worse, it’s DisplayPort 1.1, not 1.2 (except in the very latest silicon that Apple isn’t using yet).

A display manufacturer could reduce refresh rates, but 5120×2880 needs to drop to 46.335 Hz to fit in DisplayPort 1.2.

29 days ago on Samsung beats Chromebook Pixel and Retina MacBook with new high-res laptop display 1 recommend

Comment

Well, now, software supports 3840×2400 displays just fine. (Hardware’s just now catching up, though, for full refresh rate.)

I can hook my IBM T221 to my MBPR, and HiDPI works just like it does on the internal display. (I don’t do that, because the entire reason I have the T221 is to run it at native, but it can run HiDPI just fine.)

The 4096×2560 panels are for medical applications, admittedly. The 3840×2160 displays, the one I found was actually marketed for security camera applications (four 1080p images at once, or nine 720p images), although I think some are also sold for editing 4K video content.

29 days ago on Samsung beats Chromebook Pixel and Retina MacBook with new high-res laptop display 1 reply 2 recommends

Comment

Of course, you can set it to above native resolution, as BoeJob said.

And, with a small hack, you can disable HiDPI mode entirely, giving you 2048×1280, 2560×1600, 2880×1800, 3280×2100, or (on the 15.4") 3840×2400 desktop area.

29 days ago on Samsung beats Chromebook Pixel and Retina MacBook with new high-res laptop display 1 recommend

Comment

It never was king.

3840×2400 at 22.2" was out before 2560×1600 displays came out. (Although 3840×2400 did go away after 2005, the only company making those panels was dissolved.)

There’s also 4096×2560 29.6" displays, although they’re greyscale. If you want color, there’s now 3840×2160 32" displays that you can buy today.

29 days ago on Samsung beats Chromebook Pixel and Retina MacBook with new high-res laptop display 2 replies 3 recommends

Comment

One problem with high density desktop displays is the interconnect technology.

IBM did the T220 and T221 at 3840×2400, 22.2", and had tons of problems with interconnects. The T220 and early T221s needed four DVI connections just to run at 41 Hz, each running at 1920×1200 (there was enough bandwidth there for 60 Hz, though – DVI was originally specced to do 1920×1200 at 60 Hz). Later ones could do one dual-link connection at 2624×2400 @ 48 Hz, and one single-link at 1216×2400 @ 48 Hz.

DisplayPort 1.2 can help here, a 4-lane 1.2 link can do up to 17.28 Gbit/s. 4096×2560, 60 Hz, 24 bpp, reduced blanking fits in that. (Unfortunately, 5120×2880, 60 Hz, 24 bpp, reduced blanking does not.) The most that DisplayPort 1.2 can do, with a refresh rate something like 60 Hz, 24 bpp, and reduced blanking is 4480×2520 (that’s the closest one that’s evenly divisible by 8, which is an important constraint for many graphics cards, and it’s also evenly divisible by 16, which is very handy for legacy software wanting to render at 1/4 that framebuffer size), 59.866 Hz.

Also, IBM did develop DPVL, which is what DisplayPort is loosely based on, and DPVL only sent the areas of the screen that were updated. This way, a slower link can handle high framerates as long as no major changes occur. (But, that sucks for video or games.)

29 days ago on Samsung beats Chromebook Pixel and Retina MacBook with new high-res laptop display 1 reply 1 recommend

Comment

I’ll note that NHK did a study on this during their research on Super Hi-Vision, and found that 155 cycles per degree (so about 310 pixels per degree) is needed to approximate reality as closely as possible.

60 pixels per degree, or 30 cycles per degree, is the line that Retina displays cross.

As the 13.3" MBPR reaches retina at a viewing distance of 15.1" away (being 227 ppi), a display that reaches NHK’s goal would be 1173 ppi at the same viewing distance, or (for a 13.3" 16:10 display) about 13229×8268.

29 days ago on Samsung beats Chromebook Pixel and Retina MacBook with new high-res laptop display 1 reply 9 recommends