More on DisplayPort...Part 2 of 2

Part 1 of this topic is here.

While the case for moving forward from VGA and DVI is a fairly obvious one for many of us, the logic on why we need DisplayPort in a world where HDMI (High-Definition Multimedia Interface) has taken hold, may be more subtle.

HDMI has become the standard for high definition television displays and, by extension, the devices that connect to them.Its ability to support multiple audio channels, nearly any video or computer display format and as of version 1.4, an option for 100 Mbit/s Ethernet connectivity would seem to make HDMI a clear contender as the omnipotent display connection choice for all entertainment and data display applications.
So what is gained from adding DisplayPort to the landscape?

In the list of advantages over DVI, both HDMI and DisplayPort can carry audio and each has the ability to use RGB or Y’CbCr colorspace (VGA and DVI are RGB only).

There are several reasons why DisplayPort may be better in certain circumstances…and in cases where there are several reasons for anything, one of them is often ‘money.’

In our scenario, the cost factor referenced most by manufacturers is HDMI’s licensing fees. The cost of licensing HDMI in the PC display space is apparently not as practical as it is for the consumer television market. DisplayPort is a royalty-free, VESA (Visual Electronics Standards Association)-defined standard.

Another application that makes DisplayPort technology attractive is “chip-to-chip” interface for use inside a device that has an integrated display (think laptops and smartphones, currently using low voltage differential signaling or LVDS), as well as a “box-to-box” for connecting external displays. This creates interesting opportunities down the road for external displays to become lighter and thinner (and less expensive) by jettisoning the considerable electronics dedicated to scaling and other “receive signal and deploy pixels” sort of duties inside the display and making the display “direct drive.” Manufacturers can also cut costs by standardizing on one method of driving integrated and external displays. HDMI is designed as a “box-to-box” connection only.

As our requirements for computer display performance continue to expand, ideally our next connectivity standard would be able to grow as well. HDMI has a lot of advantages over DVI, but one limitation the two share is having an external clock. This limits the ultimate speed and bandwidth of the pipeline to the predetermined maximum rates already set in the architecture. In a case like this, the standard needs to be revised to extend the capabilities of the protocol as in the case of HDMI 1.3 increasing the clock speed to 340 MHz over the 165 MHz in HDMI 1.2 to enable support of WQXGA displays (the 2560x1600 of 30” LCDs most typically) . DisplayPort embeds its clock in the data signal itself, making it scalable, along with data payloads, to the physical limits of the pipeline.

So…we know some of the advantages of DisplayPort…what are the limitations?

First, HDMI is backwards compatible with DVI and you can drive an HDMI display with a DVI output. DisplayPort can be adapted and converted to HDMI or DVI, but of course the signal would have to be compatible with the destination. In other words a Y’CbCr signal could be sent through an adapter from DisplayPort to HDMI, but DVI can only handle RGB.

Second, HDMI supports xvYCC or “extended-gamut YCC” whereas DisplayPort does not. xvYCC is a color space that utilizes the full gamut of RGB grayscale, which would use all values 0-255 in an 8 bit grayscale versus a typical television gamut which would confine legal values to 16-235 under BT.601 and BT.709.

Third, HDMI supports Dolby TrueHD and DTS-HD Master Audio, which is one reason why HDMI is very entrenched in consumer products. For computer displays used in post production environments, support of these formats is far less an issue.

In the real world of motion visual post production (much of it which can no longer be described as “film” and some it even awkward to designate as “video”…) , both standards have some foothold.

The HP DreamColor display has been causing many of us in the image-handling world to reconsider the configuration of our systems to be able to monitor Deep Color…in this case, 30 bit color precision (10 bits per channel, effectively a palette of 1 billion colors). The DreamColor will connect to DisplayPort or HDMI 1.3 outputs…along with DVI-Integrated. Of course, DVI will only work with 24 bit RGB signals, but it’s a clear sign that DVI’s epitaph isn’t quite written yet. (The DreamColor also has S-video and composite video inputs…a bit of a trailer hitch on a Ferrari in my mind.)

Several manufacturers have released HDMI in/out cards for use as ingest/output devices for video editors taking in material from an HDMI-enabled camcorder, and several manufacturers have added HDMI capability to their computer display cards.

AJA Video Systems recently came out with their “LHi” line of video cards, which not only features all the traditional video industry standard interfaces such as HDSDI and analog component video, but now includes HDMI in/out.

DisplayPort has been adopted in varying degrees by many other manufacturers, and has seen a commitment as the next-generation display solution in NVIDIA’s line of professional display cards and Apple Computer’s laptops, as well as a fair number of their consumer desktops.

As for myself, I do color correction work and I also do conventional post production and editing work and I see Deep Color devices and workflows as a way to gain precision in my work. HDMI will likely be a very neat and clean way to drive a television display to view output in that environment, but I look forward to the sort of technical and economic advancements that DisplayPort will enable for those of us who need a standard that will stabilize yet remain extensible.

…and who among us wouldn’t love to add just one more cable type to the rack in the closet?

www.displayport.org
www.hdmi.org

TimK

4 cores? 8 cores? How about 240 processor cores?

NVIDIA has been making some pretty heavy-duty display cards for professionals for a number of years now. I use 2 NVIDIA Quadro dual head cards to drive my 4 monitor post production workstation. The acceleration provided for visual effects preview is extremely helpful in getting more work done in less time.
With all the focus on bigger and badder CPUs in our workstations, one of the more intriguing advancements in computer muscle is happening somewhat quietly. That would be the advent of parallel processing over a much larger group of processor cores.
While I'm absolutely positive we'll continue to see advancements in CPU power, one CPU core represents the capability to process one operation at a time...at an incredible speed of course. When you add more physical processors, you gain processing power but your limitation becomes how well the math can be sectioned up between two processes and the energy expended to figure out how to divide the operations up-and on the back side, reassemble the results into one unified dataset.
When you have multiple logical cores on one physical wafer, you now have the ability to do multiple operations using each logical core, limited by the pipeline that gets the operations on and off the chip as well as the efficiency of the code to divide and reassemble data. Multiple physical processors would involve multiple "pipes" to get data on and off each processor, gaining some extra torque over an equal configuration utilizing the same number of logical cores on one physical processor.
For operations like 3d animation or complex visual effects where the data that needs to be streamed onto the processor and the math involved is small in relationship to the processing necessary, multiple physical or logical cores are of immediate benefit. In video editing applications, adding processor cores can be helpful where large amounts of decode and encode operations are necessary, say when editing highly compressed HDV or AVCHD footage. In applications where the material is less compressed, or even uncompressed, multiple processor cores become less of an advantage as the dataset that needs to be moved becomes larger, but requires less processing, moving the speed burden to hard drives and buss speed.
NVIDIA has recently started focusing on their CUDA technology. CUDA is what gives software manufacturers a way to tap into the processing architecture of NVIDIA's powerful graphics cards to complete processes that may or may not be display or graphics related. NVIDIA uses parallel processing to get the speed from their configuration. While the cores may be smaller, there are a LOT more of them. The Quadro 5800 card for instance, has 240 processor cores. One example of utilization of this kind of processing is the CUDA-enabled RapiHD™ H.264 encoding plug-in for NVIDIA Quadro cards.
(Wikipedia's take on parallel processing-good general info.)
A way to picture the relative capability of an 8 core CPU and a 240 core GPU might be to picture five decks of cards being dealt out. With eight dealers, each dealer has 32.5 cards to distribute...with 240 dealers, each one distributes 1.083 cards. Even when we take into account that the 240 processor cores are smaller, the share of the load they have to carry is MUCH smaller and the processing is all happening at the same time. The 8 dealers may be very fast but they can't throw 32 cards out at the same time and expect them to fall neatly in front of each player in the proper rotation...they have to go one at a time. They may be dealing cards out of a pitching machine at a velocity that could severe a human limb, but the cards still have to be handled one at a time-serially. In the case of the 240 dealers in parallel, they also handle the cards one at a time with the one caveat that all but 20 of them are only handling one card with one destination. Which way do you think would be faster?
I think that GPU based processing is one of the most interesting areas of computer processing to keep an eye on... With GPUs becoming available to handle instructions along with ever more powerful CPUs, I don't think that the exponential growth in computing speed and power will be leveling off anytime soon. This technology is even being deployed as the primary processor in specialized workstations...learn more about Tesla here.

TimK

The HP Dreamcolor Display, 30 bit color...

The HP Dreamcolor monitors have been creating a bit of buzz around the post production industry. These displays utilize an LED backlight to increase the color precision of the monitor to 30 bits. (10 bits/channel as opposed to 8 bits per channel, which is how most cold cathode fluorescent backlit LCD displays seem to be defined.)

(See the blog post from December 11th, linking to Martin Euredjian's assessment of LED backlight deployment in LCD panels for another perspective.)

If DreamWorks finds the technology acceptable for their work, it would seem that it probably achieves its claims, but extensive testing is still happening and as with most new products these days, the monitor itself has had some software upgrades to improve performance...

Below, the Hard Forum Review seems pretty thorough, though the reviewer mentions he doesn't have all the hard measurement tools he'd like to substantiate a few subjective analyses...but still considerable work put in and some off-brochure info... Dithering seems to be something of concern.


The Hard Forum Review

An HP person chimes in on the Creative Cow thread below. Someone mentions the dithering mentioned in the Hard Forum review and Dan Bennett from HP says he's seen it but there's a firmware upgrade for the monitor, though the Hard Forum reviewer seems to show he has the latest firmware...

The Creative Cow.net Apple Color Forum thread

HP's FAQ on the Dreamcolor monitor is here:

The HP Dreamcolor FAQ

For reference: PNY NVIDIA's two 30 bit display cards capable of driving the Dreamcolor:

QuadroFX 4800 (one Dual-link DVI port)

QuadroFX 5800 (two dual-link DVI ports)

There are other display card vendors of course, and I've tried to scan the product lines of ATI and Matrox for 30 bit capable cards...while ATI appears to have some products with 30 bit support on PC only (24 bit on Mac), it's difficult to discern if Matrox has a product or not. I'm almost positive I've encountered a description of a Matrox product that supports 30 bit...I'm checking further.

(Update: After a few email exchanges with Matrox, it appears that the instances where Matrox display cards talk about display cards with numbers larger than 24 bit, they are referring to 32 bit, which would be 3-8 bit color channels an 8 bit alpha channel, not 30 bit color precision, which would be 3-10 bit color channels.)


TimK