II: Ingredients of the Faceplate

Frame proportions

All seven HAL faceplates that appear in the film have the same precise proportions and sizes, relative to the lens diameter. They also match the Ordway/Johnson blueprint proportions exactly.

I stress this fact because there’s a persistent belief that there were variants in the screen-seen HAL faceplates, when a simple check of available screenshots shows that there weren’t.

The Randall/Jackson prop’s aspect ratio is not seen in the film, or in any behind the scenes photos that I’ve examined.

Incidentally, sometimes people try to draw comparisons between the monolith and the HAL 9000 faceplate shape, and excitedly pull in the 1:4:9 ratio. Of course, HAL does not match the monolith’s ratio at all, and it should be noted that the 1:4:9 ratio was a later concept and embellishment for the novelization by Arthur C. Clarke — the film monoliths were of different ratios, and did not follow 1:4:9. (“The 1 x 22 x 32 was my gimmick later“ – Clarke quoted by Agel.)

The faceplate

So that’s the shape. What was the faceplate made from? Well, it was basically a Nikkor (Nikon) fisheye lens attached to a metal plate. Here are the details of the non-lens parts of the frame. The lens warrants its own section – chapter III.

The frame

Each faceplate’s outer frame was made from four bare aluminium bars, mitred 45° on each corner. No screwheads or other fasteners are visible in the film. According to the blueprint reproduced in Adam Johnson’s Lost Science, the frame was of strangely arbitrary dimensions: 13 5/8" tall and 4 9/16" wide. That’s 346mm by 116mm, but since the prop was built in Britain in the 1960s it was not made to metric specs. All metric measurements quoted here, at least regarding the HAL prop and not the Japanese-made lens, are therefore conversions. The bar material for both the outer frame and the crossbar was 1/8" or 3.18mm wide.

Given that the Nikon fisheye lens used has a diameter of 82mm at the widest point (the grip), these blueprint proportions and dimensions correspond to the 4K photo pretty precisely.

Based on this data I’ve put together the following faceplate dimensions diagram.

Interestingly, the Randall/Jackson prop has a metal back that’s clearly machined out of a solid piece of aluminium, with a tapered section for the rear of the lens socket. The outer frame also extends out from this back piece, and there are four drilled and tapped holes for mounting it. It seems probable that a pair of horizontal crossbars was attached to these holes via thumbscrews or something similar, making it easy for the set builders to install the faceplate into an opening in the set.

This level of solid, high-quality construction is what you’d expect from something built for 2001.

In this shot, Peter Jackson is holding the Nikon lens in the frame upside-down to show how it arrived from the auctioneers. He said he was unfamiliar with the design of this unusual lens at the time, and didn't realize that the anodized black dome is the lens cap, and that the protruding black tube is part of the rear lens assembly

Bevelled or not?

Screengrabs from the film itself suggest that the top of the frame may not have been flat and square. In other words, there appears to have been a slightly bevelling inwards on the top; it wasn’t parallel to the black surface. This is clearly visible from the shadowing of the frame pieces in many photos, including the two images below.

This bevel or chamfer is rarely reproduced in prop replicas. I believe 2001 researcher Karl Tate may have been the first person to call this detail out. The angle of the bevel is not known, but it looks fairly subtle. For my replica I went for 5°, which seems to reflect what we see in the film.

The crossbar has a flat top, and seems to have lined up to the lower of the bevelled angles on the long side pieces.

Top: flat-topped vertical bars seen in cross-section. Bottom: bevelled inwards by a few degrees.

However, not everyone has arrived at this conclusion. Adam Johnson believes that the frame tops were flat, not bevelled. I don’t know for sure, but the Randall/Jackson prop does not appear to be bevelled, based on the Adam Savage/Tested video.

Frame height/thickness

HAL’s faceplates are always embedded in wall or control console surfaces. The frame never sticks out more than maybe 3/32" or so – about 2.4mm. The protruding back of the lens was thus sunk into, and concealed by, whatever the faceplate was mounted to.

Peter Jackson and Adam Savage inspect the HAL prototype plate.

The Randall/Jackson prop’s visible frame included a lip extending past the back of the panel – there was an overhang all the way around the frame.

Did the film-used props have such an overhang? We don’t know. It does, however, seem like a logical design. A lip or overhang to the frame means that the set builders had a little wiggle room when cutting the opening for the HAL prop – the overhang would conceal any inadvertent gaps.

Top: cross-section of a frame with an overhang or lip. Bottom: a frame that would need to be inserted into a cleanly cut panel hole.

Replicas have tended to make the frame much deeper – maybe an inch or so. This is presumably for convenience in installing the lens or simulated lens.

Unfortunately this is one area where specifics are not known. We don’t have side elevation views of the faceplate in the blueprints, and the photos we have aren’t high-resolution enough to make out precise details.

The top plate

The main surface of the HAL frame, into which the lens is mounted, was probably a sheet of aluminium. Indeed, the Ordway/Johnson blueprints indicate that it was to be made of “1/8" satin black alum”. And generally speaking a satin finish has a very subtle texture and is not brushed.

So what were the movie prop’s surfaces like? There are definitely views in the film which imply a sort of linear smudging pattern that's parallel to the long edge of the frame. Consider the image below of the main centrifuge HAL. This is a stacked and mathematically averaged photo that I assembled, consisting of 20 frames from the film. The process reduces film grain and artefacting. If you look closely there is no linear texturing to the background panel. But the HAL panel itself has smudged lines, all parallel to the long edge, suggesting possible grain.

Faint parallel lines in this stacked image.

The clearest view of the faceplate is in the brain room corridor behind the cockpit. This also shows a faint series of vertical brush marks, suggestive of black anodized brushed aluminium. You can clearly see the tooling marks on the aluminium components.

The Christies photo of the Randall/Jackson prop clearly shows a similar texture. Some people have suggested that it was a finely textured hardwood painted black, but I think that’s unlikely. Woodgrain tends to have these slightly deeper grain lines in them from time to time, or sometimes tiny holes.

The damage to the thin HAL decal is clearly seen in this photo of the Randall/Jackson prop. It’s also clearly turquoise, which it isn’t in the film. So either the colour has yellowed since it was printed, or they decided to alter it for the finished props.

That said, a strong argument against the brushed and anodized black theory is that the movie props do not have prominent lateral reflections in them, as you’d expect from semi-shiny brushed material. As a matter of fact, you never see any reflections at all off the top plate in the film, suggesting they were lightly painted matte or satin black. This would explain why the texture is so difficult to see.

The Ordway/Johnson blueprint doesn’t show specific measurements for the plate surface, but you can work out from the provided dimensions that it was supposed to be 4 5/16" (109.5mm) wide by 10 7/16" (265mm) tall.

The logotype

A frequently misdrawn part of the HAL faceplate design is the HAL 9000 logotype itself. I've written up a separate page on what the actual typeface was, and how the logotype was created.

The speaker grille

The grille is another area of contention when it comes to HAL’s faceplate design, simply because we don’t have any really close-up views of it. All iterations of HAL seem to have a rectangular dark grey plate, perforated with hundreds of small holes. The devil, however, is in the details.

The best view we have is from a 4K screen grab of the centrifuge’s HAL 9000. The image below is a set of 32 individual frames of 4K footage, that I stacked and averaged mathematically to reduce noise. The result was then colour-adjusted and sharpened. Notice also how the brightness of the frame tops clearly looks like a bevelled surface, illuminated from a light source slightly higher than the centre of the plate, was used.

This view seems to show a repeated horizontal groove pattern, like corrugation, that was pressed into the grille’s surface. This suggests that the piece may well have been an actual speaker grille or similar component from a real 1960s manufactured object – maybe a radio or walkie-talkie. The idea behind the corrugations was probably to make the grille sturdier but still made of thin sheet metal. However, despite best efforts nobody, to my knowledge, has found an actual product with exact grille details that match. It’s still entirely possible that it was made for the film, but that seems like a lot of work for a tiny detail.

The Ordway/Johnson blueprint indicates that there should be 18 holes vertically and 22 horizontally, for a total of 396 perforations. That matches the HAL prop in the 4K photo exactly, as well as the pod bay unit. However the blueprint lists hole diameters (3/32" or 2.38mm) and spacing (7/32" or 5.5mm) which don’t actually add up to the available space – the only significant discrepancy between the blueprint and the original props that I’ve seen. It also makes no indication of surface treatment or patterns to the grille surface, nor does it say where the grille is from.

The 4K photo seems to suggest that the holes are roughly 1.6mm (approximately 0.063", or about 1/16") in diameter, with a spacing of 5.1mm horizontally on centres, and 4.0mm vertically. Note how the vertical spacing is tighter than the horizontal: the holes are on a rectangular grid.

As noted earlier, the Randall/Jackson prop has a much shorter grille than any of the HALs seen in the film. The artefact also has a flat black grille with much smaller, more numerous, and more tightly spaced holes than the ones in the 4K photo. However, the auction description for the prop does describe the grille as “replaced,” indicating that the auctioneers knew it was not the original part. The Randall/Jackson grille has evenly spaced holes on a square grid, suggesting that it’s a pre-drilled Veroboard electronics board painted black, unlike the rectangular grid seen in the photos.

The Randall/Jackson prop’s grille differs significantly from the ones in the actual film and is therefore not a useful data point.

The Ordway/Johnson blueprint doesn’t specify the grille’s overall dimensions, but you can work out from the other dimensions that it was supposed to be 4 5/16" (109.5mm) wide and 2 13/16" (71.4mm) tall. That matches the 4K photo exactly.

The speaker grille colour and pattern

Colour, or brightness, of the two panels is a problem. In some photos/footage the grille looks almost black, and in others it appears medium grey. However, the same faceplate can appear radically different from shot to shot and photo to photo. In fact, in some views (eg: the brain room) the colours appear reversed, with the main panel a lighter grey than the grille itself. Therefore this apparent shift in colour or brightness is definitely the result of different lighting conditions and angles of view.

Nonetheless the grille usually appears to be medium-dark grey; darker than gunmetal. It is slightly lighter in tone than the main panel’s solid black. It definitely wasn’t bare aluminium, like some prop replicas have it.

Both the centrifuge and bridge grilles clearly have corrugations in the highest resolution view, but look virtually flat in other scenes. This gives credence to the theory that all the grilles were corrugated, but that the corrugations were quite low in height and thus only visible close up.

Both the photo above, and the pod bay photo reproduced in “the Nikon fisheye” section earlier, suggest that the grille surface is just a tiny bit higher than the main black plate. This is consistent with the idea that the grille was, in fact, a thin sheet of metal that was glued to the plate surface.

I’ve done numerous 3D printed grilles as tests to try and figure out what the height of the “waves” was. From what I can tell the grille was very nearly flat. If you have noticeable waves – even 0.5mm to 1mm in height – then you get really pronounced horizontal reflections of any light source. This results in a high-contrast stripe effect, and we clearly don’t see striped grilles in any of the photos of the faceplate. From what I can tell the grille waves were probably around 0.2mm in height, which is absolutely negligible!

Here’s a series of test prints with the “waves” at different heights. Note how the higher waves result in obvious striping in reflected areas, whereas lower waves more closely resemble the actual prop. The lower right test print was a mere 0.2mm tall.

The waves were also quite even, like a smooth flattened sine wave. Some replicas have straight ridges running along the top of each wave crest, but that doesn’t look anything like any of the photos of the actual props.

The lens ring

A custom-machined aluminium ring was used to mount the lens on the panel. The visible part of this ring is a simple turned metal piece, with a top surface parallel to the top plate. However, set photos (such as the one in the “How many HAL faceplates were made?” section below) indicate that at least one prop was actually more complicated than that. There appears to be a deep and stepped recess designed to accommodate the ring. Did the HAL panel in that photo have an internal Nikon F lens mount so that the lens could twist in firmly? Or was it held in by friction or some other means? We may never know.

The lower edge of the ring was positioned 3 7/8" from the lower edge of the frame, according to the Ordway/Johnson blueprint. This corresponds exactly to the 4K photo. Accompanying the blueprint is a sketch of the metal mounting ring for the camera lens. According to Johnson, the dimensions of this ring (3.795" or 96.4mm in diameter; inner hole 3.094" or 78.6mm in diameter) were taken from an original artefact. It definitely would have been a crazy tight fit: Karl Tate reports that the Nikon 8mm f/8’s base ring beneath the knurled focus ring is 78.56mm in diameter.

Examination of the Adam Savage/Tested video shows that Randall/Jackson artefact has a ring with an internal surface that’s lower than top plate. This means that the lens is recessed a fair distance in. I’ve done some tests with 3D models, comparing them to still photos and screenshots, and it does seem that the top knurled ring on the movie prop lens was roughly level with the top of the lens ring. That suggests that the Randall/Jackson prop has a similar ring to the movie props, but specific measurements are not available.

You can see the ring on the pod bay lens is definitely recessed below the black plate.

Missing data

To summarize, there are several pieces of key data that are missing, making it difficult to make a wholly accurate replica of the original prop. Most are related to the fact that we don’t have side elevation blueprints, or photos of the finished prop when it wasn’t installed into the sets.

Side frame width. How much of the outer frame edge was exposed above the panel surface?

Top plate height. How much of the inner frame was exposed and visible? This appears to be a different value from the side frame width.

Inner ring depth. Exactly how deep was the interior of the aluminium ring below the surface of the black plate?

Bevelled top. Was the top surface of the frame bevelled or flat?

Lip or no lip. Was there a narrow lip around the top of the frame, making it easy to install the faceplate into a panel opening?

Side screws. Did the faceplate have rows of screws along the inner edges, as seen on the Randall/Jackson prop?

Frame depth. How deep was the entire frame assembly? (including the bit that extends behind panels)

I: Researching HAL

III: The Nikkor Lens

Contact