MT: Paul I Posted: 23 April 2014, 12:53 PM
It wasn't hard to notice there was a major problem with the print quality at this years Nationals. After all the hard work by so many people to put on a great even there must be some disappointment about this. A similar thing happened to the Relay map in Naseby a few years ago. I would imagine this could have been as a result of a wrong file type going to the printer, or something similar. I think we need a place (like this thread perhaps) where helpful facts and warnings can be shared. A go to place for all planners and controllers, in the absence of anywhere else (esp. for the inexperienced) to go.
Anyone know what happened in both these cases?
Also any of our Aussie mates have any feedback on the recent experience/quality/costs with indigo printing?
MT: Michael Posted: 23 April 2014, 1:57 PM
We never got to the bottom of “the great fuzzby problem” at Naseby. In this, OCAD 10 was used for course planning, output was to pdf, and at high magnification in Acrobat reader the map was clearly dotty whereas the course markings were perfect. The discussion is spread over pages 23-25 or so in the Mapping thread.
MT: JohnR Posted: 23 April 2014, 2:41 PM
There was a problem with the long distance elite map. It looked very much like a photo copy. Quite different to my course 8 map. Also a sharp difference between the middle course contours and the long course contours. I agree with Paul we need some work in this area. Do we have a technical committee? Counties have the Nationals next year and this will be a priority. The map printing should be the same at all events
MT: Bryan Posted: 23 April 2014, 3:54 PM
Problem was due to using complex raster image logos as background maps and having multiple embedded background maps.
We used one base master map per event which became a background map on the layout maps (several per event - one per different scale) which in turn became a background map on the final course map - useful as changes only need to be made in one place - for Ocad 10 unfortunately this meant for the sprint/long that the base map was exported as raster not vector (I was not aware of this beforehand and not picked up due to time issues - maps picked up from printers a day before first event - I picked up one issue with the middle map which required 5 trips (22km each time) to the printers and two reprints of two courses - for future major events up to nationals level I recommend closing date at least 2 weeks before first event)(and 4-5 weeks for an international event).
59 PDFs went to the printers - the relay mixed long had one PDF with 240 pages printed duplex.
Same printer as for Oceania/World Cup 2013 but different procedure used where logos were not drawn by hand as 2013 nationals logo (and RNZAF logo) too time consuming to draw by hand in vector format. I also had a colour photo on the night event as a background image.
Saving as PDF in Ocad - warning bells should be raised when the prompt comes up to enter the raster resolution - I thought it was only for the logos (and I used 400dpi - maybe 600dpi would have been better but file sizes might have become too large) but because of the embedded background maps the rasterising was also used for the base map.
For the future, a QA of each final PDF before it goes to the printer and checking the quality of the contours will show if the clarity is what is required. This should be checked by all involved - the planner, controller and person involved with the printing.
Slightly different procedure used for middle and night (where base map was imported into every layout map - a lot more time consuming especially as middle had 21 courses, 4 different layouts). Relay PDF's were created correctly in Condes 9 due to a duplex printing requirement (course and race number printed on the back of the Teslin waterproof paper saving lots of time).
Are embedded background maps still an issue with Ocad 11? I haven't played with layer management in Ocad 11 yet.
Difficult to work with different versions of Ocad (I had to worked with 8, 9 and 10 and some planners had 11) and printing courses / final maps with both Ocad and Condes (even the new version 9) is not easy (eg ensuring the coordinate system is the same between Condes and Ocad - we had three systems - New Zealand Map Grid, New Zealand Transverse Mercator and Cartesian).
Another problem with 8,9 is that the purple colour is defined as Magenta and needs redefining (I knew this intimately from Oceania).
Another tricky issue is getting the control circle sizes correct for the course/scale. Course 2 and 2a on the long had the same course but were printed at 1:15000 and 1:10000. Requires a separate export/layout map and checking that the circles are 6mm and the control descriptions are correct (eg a western depression may change to a middle depression at a different scale).
We used scales of 1:3000, 1:4000, 1:5000, 1:7500, 1:10000 and 1:15000 during the weekend.
And don't talk to me about symbol sizes and handling different symbol sets in Ocad - it's a real mess. For example, just trying to manage a night event printed at 1:4000 and 1:7500 is a nightmare where changing the symbol sizes (eg bridges, earth banks) required a complete check of the map to see if features needed redrawing - some tracks near bridges warped, some banks warped, and control descriptions changed due to the scale).
Anyone tried Open Orienteering / Purple pen at a major event?
MT: MikeB Posted: 23 April 2014, 5:27 PM
Bryan sounds like the use of raster image backgrounds should be avoided like the plague in any circumstances.
Where there any raster images in the middle or relay maps because the ones I saw were fine, it was just the sprint and long maps where I noticed it.
I had a similar issue with them a few years ago on a forest map for the Schools Champs and the contours looked very similar to those on the long maps. Fuzzy and hard to read. My dpi was even slightly lower than the one you used, 300 I think.
The printing was done the day before and by the time it was noticed it was too late to reprint so I just went with it. Thankfully most of the kids didn't realise but with a major event there are too many experienced people competing to get away with it not being noticed.
Something for CM to be aware for next year.
Did you run any test prints prior to final map printing?
At the nationals a couple of year's ago we did exactly that. so we could decide on colour settings, the right grade paper and whether it was to glossy not glossy enough and so on.
It ended up being more of a decision by the setters and controllers to what we thought was the best. Hopefully at that stage, if there are any problems, they can be sorted.
It sounds like you probably needed a committee of people to keep tabs on what was happening with all the various map logos, sizing etc.
MT: Paul I Posted: 23 April 2014, 7:46 PM
Thanks Ken you really are a wealth of ocad and printing information. Cheers for sharing your knowledge. I'm looking forward to your printing news! I suspect then Michael that the Naseby problem may have been caused in acrobat as Ken has highlighted, and possibly the pdf not generated at the highest setting. Mike I'm pretty certain our Nationals maps were rasterised because of all the logos, can you recall the final settings. I can't go back to my old emails. Or perhaps Marquita can remember.?.
MT: Marquita G Posted: 23 April 2014, 10:22 PM
No, our Nationals maps were not rasterised. Without fully understanding why, but being aware there was some sort of problem, I created the course maps in a rather roundabout way. (Michael will have kittens reading this). I used the ocad course setting software to create the courses but instead of using the “export course maps” function to create the final course maps, I turned off the background image and then copied and pasted the purple stuff (circles, lines, numbers etc) into a “real” ocad map file. A process fraught with potential danger and not one to be undertaken late at night, but quite doable with a clear head and a bit of care. (When I did the MTBO carnival maps earlier this year, this process was the only way I could figure out how to get circles and lines to be in lower purple i.e below black). Somehow this seemed to avoid the issue of raster images dragging the print quality down to the lowest common denominator. I'm sure the files we sent to the printer were PDFs of these ocad files.
MT: Bryan Posted: 24 April 2014, 8:48 AM
Well done Theoman - you obviously were well organised with plenty of time. We had adequate time for Oceania with the result that there were few problems with enough time to fix issues. Best to draw all logos by hand if you can (but usually not as good as original logo - especially for corporate logos) and avoid raster images.
Some more comments: - background Ocad files in final courses files should be avoided like the plague. You know that everything will be vectorised when the question to enter the raster quality does NOT come up when exporting the PDF. This is not an elegant solution and it is time consuming creating a final course file containing purple course file, base map cut area imported second, with layout map imported last - this can result in a zillion symbols and a zillion colours if you have lots of logos. For logo Ocad files I delete every unused symbol and every unused colour.
Also a real pain when every planner/controller wants to make last second changes to the maps on every event (as happened at Oceania).
For the nationals, the sprint and long used background maps, resulting in poor quality, for the middle and night the base map was imported directly into the final Ocad file (where quality was fine). The relay used Condes 9 (which was fine) where PDFs were rasterised with higher resolution.
- overprinting is the required effect for IOF maps - purple should not appear under black. See Ken's excellent article to achieve this: http://www.orienteering.asn.au/gfolder/mapping/Achieving-Overprint-in-Digital-Print-Maps.pdf
- The printers looked at me strangely when I mentioned possibly giving them EPS files for Oceania. They looked at me very strangely when I gave them 5 export PDFs for the offset maps for Oceania - I was annoyed that the World Champs 2013 used CMYK offset printing as I thought the IOF did not allow this (apart from getting a special dispensation). CMYK offset printing means that you can now give the printers just one PDF. Use the Export PDF in Ocad - don't use the EPS export - it is obsolete. - I couldn't find a way in Condes 9 to have a translucent logo georegistered and scaled to the border of the map - I wanted to have a 3-D effect on the border of the map with the logo. You can do this in Ocad. Look at the relay map and the logo is not on the border of the map.
MT: Michael Posted: 24 April 2014, 2:27 PM
Bryan I'm surprised you feel the middle and night were fine. My copies were quite readable but I suspect because I had the larger scales. Under a hand magnifier the edges of the lines are ragged, more so than some prints from our usual printer in the Hutt Valley. Which is the same printer WOC usually uses, is it not? Could be the paper though, what did you use?
MT: Linley Posted: 25 April 2014, 4:52 PM
I've been following this thread with interest although it is very technical. PAPO's 2014 summer of orienteering only had one map that was difficult to read. Our club policy of using Condes has served us well, as has the discipline of getting a test print before we do the run and always having the same printer on the helm. Our one glitch came about when our faithful printer had an accident and was off work.
MT: rhigham Posted: 27 April 2014, 6:03 PM
Time for me to comment Background Ocad layers do not cause printing issues - I have used up to 3 with no problem - 1) map base, 2)legend border, logos (all vectorised) then 3) course file on top. The issues arise (and always have) when a bitmap image is used somewhere in the mix. When using OCAD 10 an alert about raster image appears when exporting to PDF where a background bitmap image is used. It seems to mean all layers below the course file layer are rasterised, when this happens I usually choose at least 600 dpi - simple. However this is only a compromise as the quality is still not as good as layers that are all vector, or see below. Now when there is only one background layer (or bitmap image layer) with course and map combined on top ( archaic, inflexible and time consuming) there are no printing problems. The maps that Bryan produced had at least 2 jpg images in the background of the Nationals border layer - the Nationals logo, and the O' NZ logo -this was the issue. Bryan picked up the DPI issue in Condes but not OCAD. Looking at the PDFs under high magnification gives a slight indication something is not right, however the only way to truly analyse quality is do test prints, from the printer of the final product.
The Elephant in the room. The issues of preparing maps for a series of important events, from different planners, different scales, different course setting software, different courses, different printing processes, different image types, different georeferencing, and a distant printer is fraught and requires a specialist with plenty of time and no other commitments. Dare I say it - maybe we need a professional dedicated to stitching everything together in a short space of time, and testing stuff as they go.
Bryan works full time, was the controller for both the sprint, the night, and the WRE adviser for the long. He had issues with the WOC committee wanting to change the design for his map layout. I have the greatest respect for Bryan, and I have seen him work himself to the bone to get stuff done for major events, but we really need to think seriously about limiting amount of voluntary effort one person gives - or shit will happen.
MT: Michael Posted: 12 May 2014, 5:59 PM
Bitmap backgrounds in OCAD have been given a bad rap. That assessment may be too simplistic. I have half a dozen logos (png files) in the background of a current map, have used the OCAD pdf export, and the map itself appears to remain in vector form. That is, I cannot see any jaggies at high magnification in Adobe Reader, nor can I see the slightest suggestion of them when printed by a commercial printer. When OCAD asks for a resolution, I think it applies only to the bitmaps.
The answer may be different if the map has a background which has a background which has a background…. but there's no need to avoid bitmaps in this simple case.
MT: Michael Posted: 13 May 2014, 9:01 AM
Ken Dowling (Oz OCAD agent) says “Actually I think I did encounter it a couple of years ago when OCAD had a bug that caused rasterisation of the whole map or something - but I think that was in OCAD 10 and they fixed it promptly.” So keep your versions up to date.