Why convert images/files (scans/tifs/jpgs/bmp's or paper "flat files") into vector files or shape
Save money! Raster images or paper drawings are very costly in the long run. That's why nobody designs by hand anymore. Modifications, file retention, file sharing, data extraction and quantification of items on the drawings or maps are all easier and more accurate in digital format. Paper (flat files) and scanned images are a thing of the past, never to return, like the slide rule or vinyl records. Their time is gone.
Save time! Are you good at your business? A vectorisation service must become great at vectorisation, polygoning and web-hosting. First one becomes good, then fast. This is what they do for a living. Consequently, conversion service consumers can save time as well as money. Unless you vectorise for a living, you will not be as good or cost-effective at it as those who earn their living with it. Successful vectorisation services have developed their own specialised software that is updated and massaged on a regular basis. There are patents on some of them or aspects of these programs. Unless these routine software updates based on a constant conversion workflow are your business, you will most likely be less efficient at it than those who do this for a living. After the software conversion of the scanned image (tif, jpg , bmp etc.) comes a great deal of manual editing, which has to be fast and of high quality so that conversion service customers can focus on the creative work - and leave the 'grunt-work' of conversions to the conversion service. It follows, that such services also offer a drafting service to do updating (change orders) as well since not all conversion service customers are proficient in ordinary CAD use, or have available resources to do this work. Outsourcing is generally more cost-effective and faster than doing it yourself. Mind you, not all conversion services are created equally. Some are better than others. Caveat emptor, just like with anything else you spend money on.
Don't lose information! Each generation of dumb imaging prior to a true vector file (dwg, dgn or a shape file/GIS-ready) suffers in quality. Aperture cards and microfiche, for example, are like a photograph. Granted, for between $0.31 through $2.00 you can have the card converted to a tif image. But, it is still a 'dumb' image ('dumb' because you cannot automatically quantify anything on it, attribute or make changes, as you would when we start out with a CAD file) that has to be re-drawn the instant that modifications become necessary to the facility because nobody draws by hand anymore. Multiple generations of dumb imaging are like this: first you draw the map or drawing. Then you copy it. Then you copy it again. Then you scan it. then you print it. Then you photograph it and turn it into microfiche or some other analogue abomination. Then you scan that and put it into a pdf file. Each generation of dumb imaging is worse than the one before, and that's without the damage you might do to it through coffee stains, creases, tearing, scotch-taping, natural ageing etc. This leads to the following inescapable economics:
Fictitious Example in US Dollars: D size building drawing: to manually re-draw: 12-15 man-hours. Say 14 hours x $50.00/hour = $700.00. Whether this is done by in-house staff or an outside consultant, it's still 4.7 TIMES AS EXPENSIVE as a conversion service charging, for example $150.00 (prices may vary from one service to another and also with the density of information on the sheet, clarity, etc. There are economies of scale as well. 500 drawings or maps are likely cheaper to have converted per sheet, than one single item, for which a specification must be absorbed etc. This also frees up your draftsperson to do other (more creative) work!
Avoid Deterioration (and the art of wide format scanning): First you have a paper or transparency drawing. That necessarily deteriorates over time. With some clients, they can also have a tendency to 'grow legs' or get damaged during renovations, fires etc. This means a loss of information. Re-drawing from scratch makes this really expensive then. Add to that the various obsolete storage means: There are the aperture cards, raw scans (tifs), microfiche, etc. Then the 'dumb' image can be imported into pdf or some other format, again risking a further deterioration in quality in some cases. The quality of those 'dumb' images depends on the equipment and operator who takes the snapshot of the original drawing. Experience suggests that there is a gradient of quality. Just because you have the drawing on an aperture card or even a tif file, this most certainly does NOT mean that it's actually useful. Hopefully you've kept the paper original!
A lot depends on the shape the original is in. But even with great scanning equipment, often more than one scan is required to actually produce an image that can be converted. Resolution and brightness can have a large effect. Scanning @ 400dpi or greater is strongly recommended. Default photcopying resolutions for drawings are usually set to 200dpi. That may be fine for making a plain copy for field instructions. But that may not be enough for vectorising. Thus some folks may think they have great scans, but how useful each one is, is a very individual thing. Not all scans are equal. Some people scan at 100 dpi or less and then bellyache about the results or the messenger, such as the vectoriser who says to do it again so that it's useful. That can also be a problem because someone may have dictated the low resolution for the scans in order to save money on computers. Scanning in a lot of drawings takes space of the computer after all. Those who did that, may be uncomfortable admitting this. We're only human. Also, not all wide format scanners are created equal! Even low-end small format scanners can reach colour resolutions of 1200dpi for little money. For reasons that totally escape me, wide format colour scanning makes for a very expensive box (piece of office imaging equipment). Usually people buy combinations, printer + scanner = copier. For $30,000.00 to $60,000.00 you can get into an entry level printer/scanner combination. But if you want high resolution scanning or colour scanning, you need more money. If you want both, like you need for colour maps, bring lots more money!
Many of the boxes out there only go up to 400dpi and of course those whose limit that is will say that that is all that is required (which makes sense for plain photocopying @ 200dpi - usually a 400dpi scan can be great for vectorising too, but every now and then one encounters a nasty original that needs more...). Other scanners go up to 600 and very few go up to 1200dpi. And if you want colour with that, you better have a lot of money. Also, the software that lets the user run the scanner from the PC and also additional software for "cleaning up" scanned images, is a big factor. Unless you own that box and possibly make a living running it, you're at a disadvantage compared to the professional who does. If you think, just running the sheet though the machine is all it takes (and so it would seem when you see promotional material from the box vendors), you may be sorely mistaken in the case of poor originals in particular. And that is an understatement begging for anatomical expletives. If you vectorise for a living, own a decent and recent box and have a good handle on the software and how best to clean up images (getting rid of background junk and folding lines, coffee stains, bird droppings, etc., de-skewing photographs that were taken at an angle or de-skewing warped or twisted drawings - that's photogrammetry, its own art that good vectorisers are better off knowing), you have a distinct advantage that may become apparent in the work. Just like any computer work: "Garbage in -> Garbage out!" If you're an engineer or architect and you make your living designing or running facilities, or drawing new maps or a geologist or whatever, but your professional training and aims in life are not to specialise in the field of scanning in order to vectorise (not just scanning at some reproduction shop without concern for vectorisation at a later date), you would have a big learning curve (for which you are unlikely to have time, resources or interest). Also, a high speed wide format reproduction shop with decent equipment does not typically have any idea or interest in what it takes to scan for the purposes of vectorisation. Speed is what counts. This takes quite a bit of care and skill and art, especially with poor source documents.
The worse the original, the more care it takes to turn this into a useful image. The better the scanned image, the better the
Accuracy! Paper to CAD conversion or vectorisation services live by the accuracy of their work. It enables some to offer a warranty related to the accuracy. One and two year warranties are not entirely uncommon. Conversion service clients need to communicate clearly how they like their drawings or maps done, so that conversion personnel (not the sales rep., but the person doing the work) have crystal-clear WRITTEN directions on how to do the job. This needs to be done in writing and is clearly and obviously the responsibility of the party paying for the service.
Find the right information! Analogue (paper based) filing systems can vary in efficiency but will never match searching and finding by mouse-click. Think about the distances that contractors, your own staff and consultants have to walk to get to a flat (paper) file. Once they find it, pull it out, tear it and scotch-tape it, well, you get the idea. With web-hosting or an efficient document management system, you simply point and click, once it's vectorised. Vector files win the speed contest every time.
Avoid Duplication! Picture an older facility, which has undergone a variety of modifications. How do you make sure you're always using the latest information? You could be duplicating the wrong instructions for the right contractor. If everyone works from the same vector file, you're definitely ahead of the game. Mistakes are costly. Vector files, kept up-to-date and a superior preventive method. This is where merging of drawings or layering for different date lines, all of which are options with decent vectorisation services, become attractive.
- Standardise! If you drew an ashtray, and your uncle Gus drew that same ashtray, the drawings would look different. Likely you will prefer your drawing. Drafting is an individual art. Particularly for existing facilities with a history of modifications, it can save you a lot of time and aggravation, to use a conversion service and ask that service to stick to one standard way of doing things. Better conversion services specialise in bringing uniformity to your drawings, the way you prefer, using your preferred layering, line weights, attributes, etc. Often this communicative process creates standards for each client. Some conversion service clients already have standards on how to draw their drawings or maps and they actually stick to them. Others deviate from them somewhat and some deviate lots. Better conversion services keep records of client preferences to maintain consistency. That way, finding what you want is quick and fast and looking at the vector file does not hurt your eyes.
Why convert instead of re-draw? Conversions are ~1/5 the cost + you get a warranty with conversions. At times scanner operators wind up having to scan an image up to 6 times. A little lighter, a little darker, etc. Combine the resulting images and wind up with something that can be converted. Each generation of 'dumb imaging' (like from flat file to aperture card to tif to pdf) makes the final product more difficult. First you take a picture. Then you scan that. Then you print that. It's like using an old analogue photocopier and making a copy of a copy of a copy. The best bet is to get as good a scan as possible from the original drawing and then to convert it, so that it is in fact useful. Well equipped conversion services have high end scanners up to 1200 dpi and are experts at photogrammetry as well as clean-up of scanned images beyond the software confines that come with the original box. So if you can't clean a really tricky image and neither can any reproduction service in your area, a decent vectoriser can get more out of it than you can. You don't pay unless you're entirely happy with the work - but please, let's have clarity on your specifications first (line weights, layering preferences and other wishes). Convertors need to know the ground rules and then stick to them. In the process of determining your specifications and your exceptions to those, both buyer and vendor tend to learn and sometimes resolve certain grey areas that may have existed before. That's normal. In the end, an accurate standard for your specific requirements results. Here's an ounce of prevention for you: Customers often aren't sure about their spec's. Some change the requirements as they go along, causing more costs for the vectoriser, who is expected to just put up with it. Everyone has to make a living. Don't change the rules of the game and expect to maintain the same costs! Some folks even submit simple drawings, get a quote and then submit much busier drawings for production. How gauche.
Do all drawings have to be converted? No. If you're going to demolish a building, you won't be in any particular need to have any imaging, vector or otherwise. But if you're keeping it, and you have contractors in there to make modifications or repairs now and then (as all buildings do) then you will have a conversion done at some point. If you simply give the 'dumb' image' to a consultant (architect or engineer), then nine times out of ten, you'll be paying the $700.00 per sheet, at least, but it will be buried as part of a construction contract. Buried or not, you're losing $550.00 per sheet, for no gain whatsoever.
What about your own drafting staff? Conversion services do not exist to displace your employees. If you have people on staff exclusively to trace lines and convert from raster to vector, out-sourcing can allow you to make reductions in over-head because those people would still not be as cost-efficient as the professionals in the field. However, this is rarely the case. If you'd like your staff to be more efficient because they're doing more creative work instead of boring and time-consuming line tracing, a qualified conversion service can help. Even if you're using entry level personnel or students, young relatives, a qualified conversion service we will save you time and money and give you all the advantages described herein. The notion: "Oh, we'll just hire students..", is bunk. It will cost you more time and money than you think. Would you get a really cheap triple by-pass operation in a third world country? No? Why not? Would you just take all your old maps and drawings out camping, make a fire in the woods and use that old paper for kindling? No? You still want the information, in accurate form? Then don't mess around with kids or entry-level personnel. If your own kids makes errors, as longstanding experience suggests will occurr, who will be responsible?
What about off-the-shelf conversion software? There is a variety of such software available. You can buy it on-line or by mail-order. The sales pitch sounds terrific. Here is how it works: You run your scanned image through the software. The computer finds a line of pixels. It assigns x and y co-ordinates and makes a vector line. Then the line intersects with another line. That's where your labour kicks in. Let's say you're following a pipe. It intersects with a wall. With automatic software, the computer decides whether to turn left, right or straight. Does it now go off the pipe and make two rooms out of one? In the end, you wind up doing a lot of editing. That's when faces get long. Particularly if the sales pitch of the software emphasises that there is virtually no editing work involved (complete nonsense - or at the very least routinely understated and Utopian), disappointments result. By the time you use the program CD as a beverage coaster, somebody then has to answer for the license expenditure. The computer also has no idea about your preferences for line weights, layering etc. The other type of program is semi-automatic. You follow a line, get to an intersection and the program then cues you for a decision, for each line intersection. Maps are the most labour-intensive. You need closed polygons, not dangling lines and spaghetti (like a closed line around a forest or a property so it can be attributed, filled with a certain colour, etc.). While there are some who have the time and money to do all this editing work, a lot of folks wind up discarding the software in utter frustration and returning to manual tracing or re-drawing from scratch. Apart from that, you will not beat the labour rates of a good conversion service. Money talks. This fragmented little market has become so competitive, despite the small size of the market when looking not at the total potential number of drawings, but those hard-to-find-few folks who are prepared to spend money on conversions, that a lot of the actual 'grunt-work', meaning your first stage of editing after the raw computer conversion, is done in third world countries. Shocked? It's market demand for ultra-low pricing that drove this. The next stage is the final edit, after the file comes back from Kerplakistanville. This is preferably done by domestic operators who preferably speak the language of the drawing and who understand what the drawing is about. But being that most of the work was done cheaply, and the final edit is out of your hands when you're outsourcing, you cannot typically touch the price/performance ratio of qualified outsourcing. This is exactly where the men are separated from the boys, when it comes to vectorisation services. A superior vectorisation service has its own software. They will all say that they do but it's not always so. Also, if they do not employ and real professional engineers, architects etc. (it is common for folks to boast that they do employ professionals - in which case you might want to threaten to drop in for a visit) then you're running a greater risk for misinterpretations of the originals and a lower likelihood of getting it fixed. However, a better vectorisation service has written enough macros to speed up the work enough to be able to pay non-third-world personnel a decent wage to do the toil and still be competitive. This is very rare and highly desirable, particularly if security is an issue. Third-world countries are breeding grounds for hostile intelligence services. Expecting sub-contractors say in India or China to tow the line is a judgment call for all concerned. Outsourcing to people who use third world services, for complex drawings may then involve three parties who work on the subject files: third world vectorisers, western people to fix the third-world mistakes and an additional engineering or architectural firm or cartographer to make sure it's not full of errors that non-professionals might make. It's easy enough to find this out by dropping in and asking for the people's credentials. If you hear a lot of vagueness or references to security, compartmentalisation for security reasons or blather that they have different offices all over the place (no vectoriser is big enough to have facilities all over the place other than totally independent reproduction houses they use for scanning and sales reps working out of their homes or dealers etc.), you then must know the score: you're being sold a bill of goods here. With the right vectorisers, it should be a matter of 'case-closed', once the conversion work is done. Not hours of editing afterwards. You should also be able to drop in and see for yourself. If they turn you away, you're outsourcing to countries proliferated by hostile entities and desparate economics who then have your drawings.
Operating a facility in a CAD or GIS environment requires the prior conversion of the drawing or map. Again, the conversion cost may be buried in a larger contract, but you're still paying too much for it, particularly if a consultant does this without making use of a cost-effective conversion service. In our experience, they convert it themselves = $550.00/sheet of your money wasted.
Operating a facility with 'dumb' images' is more costly and time consuming than doing the same thing with smart vector or GIS data. Here's an example: A qualified conversion service can attribute your drawings. Move your mouse over, say a sprinkler pump. Your mouse arrow turns into a hand. You click on it and get some choices of where to go: the manufacturer's website, the product literature, the contractor's website, the service record or someplace else in a databank. That's a smart, cost-effective way to run a place. Your diesel generator is broken, you click on it and call for a service technician who goes in with armed reconnaissance, or you go exactly where you need, to order a new one or send a note off to a purchasing agent or contract specialist who can call for tenders. Your buyer gets a copy of the dwg or dgn file in question and the contractors bidding the work have exact, current information
File Size (got infinite room on your computer?): Raster image file size is determined by the scanning resolution. If you scan @ 600x600dpi, then per in˛, you have 360,000 dots. That's a lot of information to keep or to send on-line. With a vector file, you have true lines, going from one co-ordinate to another. All the space around those true vectors is not occupied by dots. Raster images therefore command a lot more real estate on your hard drive and cause delays in file sharing, downloading, etc. A vector file only takes up a small fraction of this space.
Operating with "half-dumb" (hybrid) images: A tif (scanned) image, where you go in, use an eraser tool and then stick a few CAD vectors inside of a tif. There are some tricks to making those vectors "stick" inside the raster image, so they don't shift. Qualified conversion services can typically do that, by the way. But let's say you know how to do that too. You still can only quantify information exclusively on the vector portion of that file. Most document management systems choke on hybrid files because they're neither one thing nor another. They are a band-aid and a half-measure with only partial functionality. You can't attribute items on the drawing and you're still using very large file sizes because of all the "dots" that make up the raster image image below.
Also, when you run a facility, you usually have qualified personnel, who wind up recording additional data, which is crucial for running certain equipment. They write down critical service temperatures and other things they need. Put that information onto the dwg or dgn file and you won't lose it, as you might if the only place to write that information on is a deteriorating paper copy on site, that might get lost. Even lamination does not totally prevent reduction in image clarity. Need the info in more than one place: You'll run copies. Copies of copies of deteriorating paper. Nix Cultura. Why would you do that to yourself?
With smart vector and shape files, you can confidently quantify critical information. Square footage areas for buildings and grounds are an example. Some folks have paid for lawn-mowing and snow removal for years, by the square foot. The rate per square foot was competitive but the square footage was wrong, higher of course. Occupancy issues and even eligibility for grants or maximum building use are all tied to having accurate data.
With smart vector files you can also work more safely, especially with the electrical system. Circuits can be verified and recorded. Control in general, is best achieved with a facility run in a vector environment.
If law enforcement operations become necessary, ready access to vector files enable SWAT to wargame operations right away. Generally, security and protection issues are more easily addressed with vector files. This means that you may very well save lives by having smart, digital data. Seconds count in the event of emergencies. If a felon is holed up in your place, SWAT can get him out faster and with fewer casualties if you do your homework and become all digital.
Fire prevention officers can quickly assess code compliance using vector files. An efficient fire hazard analysis requires accurate vector drawings. In the event of a fire, you will want to file a claim for damages with your insurer. If your facility has undergone a lot of modifications or additions, each of which had to comply with the codes in place at the time you filed for a building permit, then what sort of code does the whole thing together comply with, and how do you know that? What in fact is your defence-in-depth against fire? Exactly what barriers are rated for how long and in what way? Where are fire walls versus occupancy separations or ordinary fire separations? Codes change. Without accurate vector drawings of the whole place, your insurer and possibly even the investigating authority will seek to hold you responsible for negligence. Your exclusive defence in court is due diligence. Due diligence starts with an accurate, vector-based assessment of what you have where, how long it's rated for and what shape it's in and who is maintaining what and in accordance with which standards. Municipal authorities in NAFTA tend to share a certain frustration level when it comes to their ability to enforce codes, which is why convictions in this realm tend to be coveted, particularly in the event of casualties. Equally, insurance adjusters are evaluated on the basis of their ability to deflect claims by any possible means. In the event of a fire and a claim, the worst thing to have to do is to spend a lot of time going through analogue flat files to build your case - especially if they've gone up in flames or washed away with sprinklers in the event that caused the claim to happen in the first place. If you get digital (vector files of all your building drawings, including contractor as-builts) early on, and get your AHJ (authority having jurisdiction) onboard with your defence-in-depth against fire, you will have made friends you
may really need in the event of a claim. At least you'll have a fighting chance.
Firestopping (mechanical and electrical through-penetrations and joints in fire-resistance rated wall and floor assemblies) is best recorded on your drawings, otherwise you have no way of complying with fire safety regulations whatsoever See http://www.geocities.com/astximw. Older buildings often have no firestops. That means you have no operable fire barriers at all. You can have a great 3 hour rated concrete wall, but if it has one cable penetration inside of it that is unsealed or incorrectly firestopped, the 3 hours go down to 0 minutes. This increases liability substantially. If you do have firestops, and your building undergoes modifications, your old seals are re-entered. If you have no idea what the original firestop was, then how is anyone supposed to know how to repair that seal and comply with the original UL listing, without which your entire fire barrier is inoperable? Smart vector files provide the necessary roadmap to achieve compliance and more importantly, the ability to prove your case. Generally, compliance with building regulations and standards is most easily determined with vector files.
If you SELL a facility, not having vector data on it can really lower the price. When you have a change of ownership, the local fire department instantly sees its chance to enforce the local code. This necessitates more consulting work, which takes time and money, for a fire hazard analysis etc. Everything on this web page (http://www.geocities.com/ghering2000/why_vectorise.html) is also quite potentially apparent to new buyers.
What if your original drawings are inaccurate or incomplete? You're not alone! Let's say you have no "as-builts" or for whatever reason, you originals are only 80% correct. With a conversion, at least you have a starting point. If you have to move a line or add a room or a piece of equipment, this is accomplished much faster and for less cost than if you re-draw everything. Not doing anything at all is clearly a case of false economies. Even if you measure and re-draw on the basis of new measurements, you will still need to convert and merge the mechanical and electrical drawings because you cannot measure what you cannot see, such as electrical conduit running inside of walls etc. Measuring companies are not in short supply. Good vectorisers collaborate with them to merge results so that the end-user only deals with one company.
What if you don't have any staff who can draw in CAD? Change requests are no problem for qualified conversion services. Qualified conversion services can alter a drawing as you need it, similar to a drafting service. However you must comply with local regulations in terms of facility modifications that are substantial enough to require a building permit, which necessitates a review by a local professional.
If you elect to have your drawings web-hosted, there are untold advantages that save time and money. Multiple users can see the same drawing at once and do not require a software license to view, print or redline edit. You can also track who has seen what and when.
Computer Technology in Today's Architecture Office
by Ms. Bettina Mehnert, AIA
As the clock ticks towards the next century, we architects can't help but wonder what the new millennium holds in store for architectural design. Once an esoteric and expensive tool used only by the largest firms, the CAD (Computer Aided Design) workstation now becomes an essential element of any architectural office.
While some architects fight the notion that CAD is an unpleasant necessity for doing business which is relegated to other people if possible, others might think that it is the best thing that has happened to the building design profession, providing a boost in profitability and the ability to offer new services. Depending on how CAD is implemented in your office, either of these statements could be true.
For many architects, making the transition from manual drafting to drafting software to 3D supporting architectural software hinges in part on the philosophical approach a given firm follows, the types of building design it does, and how it develops the design.
In most cases, the smartest path is to add a station or two devoted to 3D visualization. More and more clients are starting to ask for, and even expect, perspectives and rendered walk-through animations, turning the competitive advantage gained through leading-edge implementation of technology into an increasingly short-lived event.
The Internet is a good example of this phenomenon. Touted as the future of networking and collaboration, it has been the focus of incredible media attention both within the public and the profession. Here at Architects Hawaii Ltd., as at many other architectural firms, an increasing number of clients asked if we could communicate with them directly using E-mail systems, FTP (File Transfer Protocol) or other tools of the information super-highway. At the same time, we needed better tools to communicate and share information with our consultants. Today, as a direct response to these external market forces and our internal functional needs, we have implemented a leased fiber optics line to our office, and every employee in our company has direct access to the Internet.
The architectural fields in which CAD programs have made the most progress are in their design capabilities and perspective rendering.
In the area of 3D renderings, walk-throughs, fly-throughs and lighting, studies can be very effective in showing a client how the finished product will look - to the point of showing how the light will shine through the different windows in the building at different times of the day and year.
Easing the transfer into three-dimensional computer rendering, can be accomplished by developing wire frame models that are then manually rendered. This is a viable solution bridging the distance between different approaches.
On the cover of this issue we are showing The Shops at Wailea, Maui. This is a new luxury resort retail shopping center, master planned as a middle to high end shopping and fine dining center.
With a large library of finish surfaces and the ability to merge proposed designs with actual photographs, we enabled our clients to preview the design months before the ground was broken. Using a variety of different software packages, the base-wireframe model was generated in AutoCAD, while sun light, materials, water features and the sky were rendered with AutoArchitect and 3D Studio Viz, as envisioned by Alain Geronimo, a designer with Architects Hawaii. We worked closely with Nick Kaars and Associates on the color system design, using a custom designed viewable color chart to obtain the subtle desired color variations on the computer screen as well as on prints. To add photorealism Earl Ma, also a designer, created a digital "entourage" image library with scanned photos of people and trees using Adobe's Photoshop and Illustrator. By digitally separating each entity, we manipulated scale, location, and sharpness individually without affecting the others. Finally, we sketched in cast shadows for these figures freehand using a graphics tablet and digital pen.
The current development of architectural software demonstrates a major trend towards building comprehensive 3D CAD models. From CAD models, all the plans, sections, elevations, and even details can be automatically generated. Drawing coordination also will be automatic, as changes to the model are reflected in all the relevant drawings and details. Until recently, however, it was just too difficult and time-consuming to actually attempt this with real building projects.
But due to the massive increases in inexpensive micro computing power, better operating systems and architecturally specialized programs with vastly improved artificial intelligence, this idealized way of designing buildings is now moving toward realistic professional practice. 3D visualizations are a side benefit, since comprehensive models must work in 3D.
The Creative Process
Has computer technology changed the way architects think, the way we design and express creativity? The answer is a definite yes and no. I don't believe technology has any influence on the actual design creativity or the development of design solutions. It is a tool aiding the professional. As such, it has added to the traditional way architects present their ideas. Technology can add excitement. It can help designers express themselves in a variety of stimulating ways, using a diverseness of media.
Does all of this mean we have reached the end of this technology curve? No, the future continues to look very exciting and the accelerated speed of technological advances show no signs of slowing down. The constant struggle of our relatively slow-moving industry trying to keep up with fast-paced technology on which we increasingly depend, will not cease.
For those architects still waiting for a more intuitive interface, aiding in implementing design ideas without having to fight the often disliked "technology beast", there is a shimmer of hope on the horizon based on the development of the following new tools:
The Digital Cocktail Napkin, for sketch and gesture recognition, Right Tool/Right Time, a context-hypersensitive tool palette; and "Digital Clay", a design-research system for deriving computer-generated 3D models from hand-drawn sketches.
The replacement of the mouse and keyboard with a stylus and voice recognition system. This will change the whole way we think about the software and the way we integrate it into our work.
A 3D digital image capture system to input video which picks up 3D large-scale environments, and converts them into fully textured 3D models. Since most works of architecture share a physical environment with other buildings, input technologies will capture not only the designer's intentions, but the urban design context as well. User-generated CAD models, such as those proposed for new buildings or alterations, can be matched and tracked into the contextual scene for an affordable level of photorealistic environmental simulation.
Some input technology, such as the "traditional" goggles-and-glove mode of virtual reality (VR), melds the input and display environments. Multiple participants can take a virtual walk through proposed designs while interacting with each other inside a virtual space.
As reported at the 1998 Microprocessor Forum, the typical AEC workstation of 2002-2004 will have a stable operating system, 100 times faster than the current best. Your company's power users will run 1GB of RAM or better in their machines. The ideal CAD station of 2004 will combine low cost and ease of use, easily navigable documents, intuitive solid modeling with seamless 2D/3D integration, lightning speed, integrated 3D input/output, photo realism, every conceivable mode of fly-by , drive-through and walk-about plus universal communications capability over any type of medium imaginable. Considered leading-edge today, those technologies will be part of the AEC mainstream before 2004.
Cyber Architecture in Cyberspace
From the 1920s to 60s, the whole notion that architecture is about the experimental modulation of space and time - that is the "fourth dimensional" - captivated architectural theory.
Architecture seen for its symbolic content or "message system" has taken on a life of its own. Not only have architecture drawings generated art markets in their own right - as illustrated conceptual art, buildings have begun to become arguments in architectural discourse about architecture, as propositions, narratives, and inquiries that happen also to be inhabitable.
In the avant-garde guise, the movement goes by the name of Deconstructivism, or Post-Structuralism. Its interest is neither a building as an object of beauty nor inhabitation, Architecture becomes an object of information, a collection of ciphers and "moves", junctions and disjunctions, reversals and iterations, metaphorical woundings and healings, and so on, all to be "read".
Architecture, in its physical form, is no longer an important issue, the meaning or what can be read reigns supreme. Would the next step of cyber architecture in cyberspace be a logical approach. Furthermore, cyberspace can provide and do what architects have been dreaming of all the time, e.g. build a castle in the air without being constrained by gravity.
Architecture in cyberspace provides an opportunity for architects to explore and test out and idea before actually building it. It allows the user to do walkthroughs and provides a more realistic view of what it would be like in real life. It allows different professionals to interact and simulate an idea before actually building. It allows theories and concepts to be realised and explained even with a physical body and thus generates newer ideas never though possible. It is like paper architecture, except that it is much more.
Fundamentally, the idea of architecture per se as a ground for discourse and theorising works fine. But the question remains whether there is architecture in cyberspace.
Cyberspace architecture can be easily related to a physical parallel. It is like the museum in real life and a model in cyberspace. But buildings in cyberspace are constructed from programming language and not bricks and mortar.
Although there is so much hype about cyberspace in the media, how cyber can architecture be? An article recently posted in the newsgroup, alt.architecture, stated that "After all, Internet is just an electronic billboard!", the author said that users were becoming irritated by animations which took up considerable time when downloading web pages…he wrote, "Just get to the point!". This applies to cyber architecture too, if the design remains in bits and bytes, architects often face a difficult question of "So what?".