This looks very interesting to me in that it covers the whole Earth not just a sample area. Anyone know anything about this company? I might be interested in this package if it does in fact do what it claims about loading scenery father away and NOT CHANGING vwhen approaching but I suppose unless one is a beta tester it remains only a claim. Another thing that concerns me is that this seems to be limited to elevated data only, not cities and roads etc. At least it appears this way from the screenshots whereas Megascenery covers both all this pretty well except very limited in total coverage. I certainly am not interested in paying 60 US for moutains alone but like I said not sure if that is the case. Best Wishes,[h4]Randy J. Smith[/h4][h3]P M D G's 747-400[/h3][h4]coming to a runway near you[/h4][/font color]Caution! Not a real pilot, but do play one on TV ;-)AMD 64 3200+ | ASUS KV8 DELUXE | GFORCE 5700 ULTRA @535/1000 | Maxtor 6Y080M0 SATA 80 GIG | 512 DDR 400 | Windows Xp Pro | Windows Xp Pro 64 |

Not sure about this "far away" stuff - sounds like this is some MSFS thing that can probably be achieved without this product. But I am just guessing.But to me it primarily looks like the global mesh scenery (based on Space Shuttle SRTM) and frankly I don't care for having detailed mesh in South America, Africa or Asia where I am never going fly. Therefore I prefer something like FS Terrain (also based on SRTM data) but concentrates on US and Europe and probably infinitely more affordable.Michael J.http://www.reality-xp.com/community/nr/rsc/rxp-higher.jpg

I'm very excited about this release too! Just imagine a highly detailed mesh covering the whole planet! If it is as good and complete as it claims, 60 Dollars is a real bargain!I have some questions, though:1. How well are the default airports integrated into the mesh? There where many problems in the past with meshes and airports to be higher or lower than the surrounding terrain.2. Which areas are covered and in which detail (i.e LOD)? Can we expect detailed Antarktis aswell?3. How well do the default, not very detailed coastlines bend with the new mesh? Ex: Are cliffs nearby seas visible even when the coastline information would "exclude" them?cheers,Claudio

Seems to be two versions both LOD9 and LOD5 (?)For loads of screenshots and info:http://fsglobal2005.com/At least Greenland is modeled in the cold areas ;)

The Tahiti airfield area looks sweet. I would like to see other areas like SFO-KOAK-KRNO, of course I would be willing to pay for this if it at least matches Megascenery for these areas plus the added bonus of the whole world. Well I can always take the plunge and give them a chance. Scenery has always been lacking I believe in flight simulators either for distance to shimering or something else. Maybe it's just a MS limiting issue but and I hope they get the point that most of us desire high quality scenery ;DBest Wishes,[h4]Randy J. Smith[/h4][h3]P M D G's 747-400[/h3][h4]coming to a runway near you[/h4][/font color]Caution! Not a real pilot, but do play one on TV ;-)AMD 64 3200+ | ASUS KV8 DELUXE | GFORCE 5700 ULTRA @535/1000 | Maxtor 6Y080M0 SATA 80 GIG | 512 DDR 400 | Windows Xp Pro | Windows Xp Pro 64 |

The problems that you normally have with mesh will still be there with this mesh.The "plateau" or "stair-step" effect is caused by two major issues, which can not be easily corrected.1. The published altitude of the airport is either incorrect, or the placement of the airport is incorrect. This causes the airport to stick out above the surrounding landscape that is correct.2. The airport is not entirely flat. many runways are built on slight, (and in some extreme bush cases, severe) inclines. All airports in MSFS must be default be flat. In that case, the airport may "start" at the proper level, but end up in a box canyon, or on a plateau because the engine does not allow sloping runways.Lakes, shorelines, and Rivers also have similar problems.The amount of work necessary to use "flattens" to correct every anomoly is not reasonable. This product will probably be fine, in a general sense, but I would not expect it to be better than the efforts of some other mesh authors that have spent time concentrating on smaller areas, such as Justin Tyme, or Holgar Sandmann.

I like to take an SRTM data tile and make my own, experimenting with interpolation, etc.I didn't realize FS Global was $60. Still, a product doing 76m over the entire world (well, 60 deg N to 60 deg S) is a daunting task to prepare.

Howdy,here are a few of my thoughts based on the advertisement. They are not meant to "take down" the product (after all, I haven't had a chance to test it) but raise a few flags for potential users:1. The authors include a LOD5 "buffer mesh", which they term - somewhat breathlessly - "double-rendering". While this is definitely a good idea for areas for which the default mesh is only LOD5 (most of the planet outside the US), I wonder why the buffer mesh itself is made at LOD5 and not LOD6? Unless Global2005 deactivates the default terrain folder or installs itself at the very bottom of the Scenery Library, the default LOD5 mesh will be used instead, thanks to the quirky reversal of mesh priority settings in FS2004. 2. The German reviewer (link on the Global2005 website), predictably, also visted the Grand Canyon and approved of the great detail. But did he, in fact, look at the Global2005 mesh? Probably not, since the canyon is covered by a MS default LOD9 mesh, which will override any add-on LOD9 activated higher up in the Scenery Library. More generally, for the US and Alaska, the recently released freeware LOD10 files of both FSGenesis and FSFreeware will be much better than the Global2005 coverage. 3. They guarantee (!) that any add-on scenery will work with Global2005. That's a very misleading statement, at best! Any add-on scenery outside of airfield flatten areas, like most of the bush flying add-ons along lakes and ocean inlets, will not work as intended. Well, they won't crash your FS but you might need a helicopter to get to the scenery-in-a-hole. 4. The website says something about a different mathematical process to retain the detail of the source data when resampling from the 90-m base to the 76-m LOD9 grid. While this sounds like an interesting idea, and may indeed have merit, the data still has to be sent through resample.exe, of which we know very little how it interpolates the input data (other than that it looks okay). IMHO, anyone who claims to have a better product needs to provide more information than they do or, better yet, direct proof.5. All areas north and south of 60 degrees latitude seem to be based on the same or similar low-res data that MS used for the default LOD5 mesh. Thus, don't expect great detail for northern Alaska, Scandinavia, or Siberia, never mind Antarctica. 6. Like the authors state on their website: this is a terrain mesh product only!!! No changes to shorelines, roads, lake or airport elevations etc.To end on a positive note: if Global2005 is indeed not worse (maybe better) than MyWorld, FSGenesis, and other more or less global products the price tag seems quite reasonable.Cheers, Holger

Thanks Holger... Your oppinion is greatly respected in the FS community... I would like that the product be reviewed by someone who really knows about mesh development like you...

Hi,the price tag seems more than reasonable. If you buy that mesh from other vendors you'll probably get just the USA or whatever.What I like most is that I can fly everywhere now and get some decent terrain. But landclass is still missing...No more chase for freeware terrain that covers some destinations where you just want to fly. No more downloads from "All nodes are busy".com BTW: I got my copy yesterday. Fast shipping, but Germany is not far away from Austra anyway :-)I had a problem installing the mesh. The installer only copied the LOD5 buffer mesh in the installation folders. But since the *.bin files on the DVDs are just ZIP-files I copied the missing bgls over manually.Regarding the buffer mesh: I still get these cracks in the landscape though they are smaller and more far away now.Alex

No more downloads>from "All nodes are busy".com >LOL LOL LOLThanks for the review.

Friends,we are proud, that our product has risen curiosity on this forum.As we can read, some points remain in the unclear. Since we do not want to hide anything, we will improve our website soon to address these points and give even better information. Before we do so, we think, that this forum deserves a personal answer first. >1. The authors include a LOD5 "buffer mesh", which they term ->somewhat breathlessly - "double-rendering". While this is>definitely a good idea for areas for which the default mesh is>only LOD5 (most of the planet outside the US), I wonder why>the buffer mesh itself is made at LOD5 and not LOD6? Unless>Global2005 deactivates the default terrain folder or installs>itself at the very bottom of the Scenery Library, the default>LOD5 mesh will be used instead, thanks to the quirky reversal>of mesh priority settings in FS2004. This issue was extensively tested and thought about. To cut a long sentence short, we did not use LOD6, beause this is not the LOD intended by Microsoft for far distance view. LOD6 loads by far later in the SIM than LOD5 and thus would have had not the effect in the far distance we wanted. In fact it would have been completely useless.The effect of the LOD5 "Double-Render" mesh was tested and compared on various testing PCs. Especially in areas where the default LOD5 mesh looks unreal. There definitely was a change for the better, which was also confirmed by our beta testers. Pickets in the far distance vanished completely. On the other hand, it is quite easy to move layers up and down in scenery library, if one wants to test that more.>2. The German reviewer (link on the Global2005 website),>predictably, also visted the Grand Canyon and approved of the>great detail. But did he, in fact, look at the Global2005>mesh? Probably not, since the canyon is covered by a MS>default LOD9 mesh, which will override any add-on LOD9>activated higher up in the Scenery Library. More generally,>for the US and Alaska, the recently released freeware LOD10>files of both FSGenesis and FSFreeware will be much better>than the Global2005 coverage. That is perfectly right and we have on influence on the reviews. It does, however, in fact show the purpose of FSGlobal 2005. We do NOT want to be the "one and only" mesh. We DO want to provide a consistent mesh covery of the whole planet based on the latest and best available data of consistent source. We could have included a LOD10 mesh, yes. But that would have been against the philosophy of the product. That again would have left the user with different mesh qualities in different areas not to think about the datasize. We already have 12,6GB! FSGlobal 2005 is designed and waits for small local meshes in better quality. It is designed to be a base. For designers, freeware or else, to set upon and no need to include data-intensive meshes in own products. Currently w/o FSGlobal 2005 no one can programm a local airport scenery in the Andes. It simply would look unreal with the standard mesh around it. FSGlobal 2005 gives the world, especially the so far "neglected" areas a shape one doesn't need to be ashamed of.>3. They guarantee (!) that any add-on scenery will work with>Global2005. That's a very misleading statement, at best! Any>add-on scenery outside of airfield flatten areas, like most of>the bush flying add-ons along lakes and ocean inlets, will not>work as intended. Well, they won't crash your FS but you might>need a helicopter to get to the scenery-in-a-hole.Weeeell, we can't accept that like it is said here. In no way it is that absolut. It very much depends on how the scenery was programmed. If the scenery was programmed using absolute altitudes or using wrong altitudes it can (and probably will) look odd. But we consider that a bug of the "other" scenery. Of course (and we thought that is obvious and logical) FSGlobal 2005 can only work correctly with that scenery, if that scenery is programmed correctly. And we do think that using correct altitudes is one main issue for a correctly programmed detail scenery. The mesh simply can't do anything here. It is forced to these altitudes. On the other hand, most sceneries are programmed using correct altitudes or having only very small difference. This won't be noticed at all or may even look correct or good.One special problem are runways which have slopes. Most runway ends have different altitudes. FS can have only one airport elevation. FSGlobal 2005 will show the correct terrain rendering.We do think that any difficiencies are overwhelmingly outweight by the advantages.And one main thing: All these issues also apply to the other meshes!>4. The website says something about a different mathematical>process to retain the detail of the source data when>resampling from the 90-m base to the 76-m LOD9 grid. While>this sounds like an interesting idea, and may indeed have>merit, the data still has to be sent through resample.exe, of>which we know very little how it interpolates the input data>other than that it looks okay). IMHO, anyone who claims to>have a better product needs to provide more information than>they do or, better yet, direct proof.Yes, resample.exe was used. Since everyone uses it, every product undergoes the same "unknown" changes it may apply. The more important question is, WHICH data is sent to resample.exe and how this data might have been improved over normal SRTM data to provide optimal results. This is, what we mean with the difficult mathematical processes.We do not want to directly compare FSGlobal 2005 with other (party competing) mesh-products for obvious reasons. FSGlobal 2005 comes on 3 full DVDs with a datasize of 12,6 GB. What data-size do others have? We have a lot of coparing screenshots available. On purpose we did shoot them with lat/long/alt/direction showing at the top. Everyone using a competing mesh can just go to that area and check.>5. All areas north and south of 60 degrees latitude seem to be>based on the same or similar low-res data that MS used for the>default LOD5 mesh. Thus, don't expect great detail for>northern Alaska, Scandinavia, or Siberia, never mind>Antarctica. This is a point we very much "chew at" ourselves. I mean, it is a pity, that the shuttle did not measure the whole globe. However, we have to cope with that fact. NASAs reasoning was, that with leaving these areas out, still 98% of the inhabited areas of earth (and thus those with most of the aeronatical traffic) would be caught. We could have left out these areas completely as well. We did not. We used various available data (unfortunately I am not allowed to disclose the source here) and with our preprocessing we were able to improve the shape of these areas over the standard FS Mesh (of course on a lower quality base then the SRTM data areas). >6. Like the authors state on their website: this is a terrain>mesh product only!!! No changes to shorelines, roads, lake or>airport elevations etc.That is correct. It is the philosophy behind the product. We do not want to interfere with peoples personal installations and other add-on scenery. We see FSGlobal 2005 as a replacement for the standard FS Mesh. I personally do not need any other scenery any more other than that. Who wants more details (literally speaking the "gardentwarf" in the backyard) can additionally install higher detailed areas. Who doesn't, has a much better rendering of the globae than in standard FS scenery. There are landclass files of various sources available on the net and as payware. In fact FSGlobal 2005 *IS* the best mesh rendering available at that time on a global datasource.I am open for any factual discussion here.Stefan Schaefer

Hi Stefan,thanks for the prompt and detailed feedback to my points - much appreciated!A few more comments re your comments ;-)Re 1.) Maybe we're talking about two different issues here. With my newer mesh files I use a LOD7 buffer mesh to eliminate the annoying "texture tearing" (bluish slivers in the mid(!) distance) that often occurs at the transition line of add-on and default mesh. Using a buffer mesh made from the same source data reduces the problem by 90-100%. It's possible that a LOD5 mesh has the same benefit but my particular concern is that it simply won't be used by the sim unless you either disable the default mesh or otherwise ensure that the reverse priority of mesh display is taken into account. What method do you use, e.g., how do you prevent the default LOD5 mesh files from being used?Re 3.) I can show many screenshots of how add-on mesh interferes with these "outside" scenery add-ons, not particularly with your mesh but with ANY add-on mesh. And it doesn't necessarily have anything to do with fixed or non-fixed object altitudes, it's simply a matter of where the scenery is placed. In many cases, the developer picks a flattish (in the default mesh!) spot to place runways and objects. Any add-on mesh is likely to interfere with this, as it will have more drastic elevation changes. I invite you to download any of the highly popular Bush Flying Unlimited sceneries - http://bfu.avsim.net/ - and see for yourself. Your "guarantee" has to come with a very big qualifier to let it stand. BTW, what practical effect does your guarantee have for the user??? ;-) Re 4.) "We do not want to directly compare FSGlobal 2005 with other (party competing) mesh-products for obvious reasons."Well, you do just that on your website by claiming superior methods without giving proof. ;-)"FSGlobal 2005 comes on 3 full DVDs with a datasize of 12,6 GB. What data-size do others have?"Ummm... that's dangerous territory - are we talking quantity over quality? Re 5.) You don't have to disclose your data sources - always makes for interesting conspiracy theories ;-) - but claims without any evidence are not helpful for the potential customer. You don't even tell us the resolution of the non-SRTM source data, compared to the default 30-arcsec data.Re 6.) "In fact FSGlobal 2005 *IS* the best mesh rendering available at that time on a global datasource."This may well be but you (and your contributors) are the only one(s) who know whether this is true or not (and even that would require perfect knowledge of the methods the other mesh designers use) and thus a statement like that is pure "posturing" unless it can be evaluated and confirmed by reviewers and users.Cheers, Holger

Hi Alex,>What I like most is that I can fly everywhere now and get some>decent terrain. But landclass is still missing...Give a look at MyWorld2004 Landclass. It's a 39MB download and it covers the entire world; after that for a specific location you can add a more detailed class file.>BTW: I got my copy yesterday. Fast shipping, but Germany is>not far away from Austra anywayI got mine in the morning here in Italy ;-)>I had a problem installing the mesh. The installer only copied>the LOD5 buffer mesh in the installation folders. But since>the *.bin files on the DVDs are just ZIP-files I copied the>missing bgls over manually.I decided for a manual installation right away and extract all 2853 .bgl files 11,7 GB (12.602.851.781 byte) in six different folders called AFR, ASI, EUR, NAM, OCE and SAM.After that i placed six new entries in scenery.cfg just below 1) Default Terrain 2) Default Scenery at a very low level.BTW on DVD box i read "Enjoy over 16Gigabyte of the very best meshdata: why i only have 12GB???Anyway i'm overall impressed by this new addon. And i had several payware mesh installed before (LAGO and so on) but no one gave me such wonderful mountains w/o considerable frame rate hit.Just my two centsLuigi ;-)

Holger,We can start the "neverending story" here discussing for ever on ... :-) This does not make sense. So let me clarify some more and then we let the users make a final decision. Opinions start to pop up in various forums now. Even in this thread you can read them. Let these opinions be our "proof" you ask for. Re. LOD5:Our main issue about adding a low-resolution mesh is to eliminate large errors in the default LOD5 mesh, like peaks in southern New Zealand or huge errors in Patagonia. After some testing and research we came to the conclusion that LOD5 was fully sufficient to accomplish this and the best compromise between quality and performance.>> What method do you use, e.g., how do you prevent the default LOD5 mesh files from being used? :() No special methods required in this case, and we confirm that our LOD5 mesh is loaded, not the default one. Otherwise you would see the famous LOD5-peaks in New Zealand in the distance, for example.Re COMPATABILITY:Our compatability promise is based on two very simple thoughts: 1) The SRTM data provide extremely precise elevation data for allsurfaces suitable for building a runway (or even a grass strip) on. So if an airport is placed somewhere at the correct real-world elevation, there might be a barely visible error of a few feet, but the scenery will be usable in any case. A problem is that real world runways (as in the SRTM data) sometimes have a significant slope, leading to an elevation error with the always flat FS2004-runways at the edge of the flatten zone. 2) If a scenery requires very-high resolution terrain that is usually provided by LOD10-mesh, either by the add-on or the default scenery. That's why we didn't include LOD10-mesh - although we would have been able to do so for U.S. and some other territory - for compatibility reasons. In most cases at least one of these cases are fulfilled, and we don't think this is a very big qualifier. It is what a user should expect from an add-on scenery. But now - if somebody creates a fantasy scenery with airfields at fantasy elevations based on insufficient data source (like old GTOPO30 data), without providing an own elevation mesh, above mentioned thoughts are busted. Of course (and that should be a logical conclusion) we cannot guarantee anything for such type of "scenery", which is fantasy. The same also applies for the rare cases where airfields (mostly grass strips) in the default scenery were set to the wrong elevation. But that is also a problem in the plain FS, like with the Table Rock airfield (US, Oregon (?)) in the FS2002 default scenery. In any case: try some bush-flying on the "SRTM world"! It is much more fun than with the default scenery! Discover some rolling hills in Congo, the Angel falls in Venezuela and details of 5000m high Pik Carstensz in Indonesia! And compare the screenshots on our web site. If you choose FSGlobal 2005, you won't have to buy each country or continent seperately, at least. We really encourage bush flying with SRTM scenery.Re NON-SRTM DATASource data resolution ranges from 25m (Alps) to ca. 460 x 920m for arctic regions. Where we used 30-arcsec data we did some heavy preprocessing to reduce the estimated error volume of the surface in the resampled final product compared to the real world surface. The basic problem is that a surface is always flattened and blured by application of a regular mesh. But by modifying some elevation points (which usually represent average elevation for an area) it is possible to regain the characteristic shape of the surface. It is like increasing the contrast in a picture, but with a more complex mathematical background. In our test areas in Europe the resulting visual shape was much closer to reality, as well as the peak elevations. So we conclude to offer an improvement in the non-SRTM areas, like Antarctica, too. It is the best one can get out of LOD5. The same technique was used for LOD9 data but the resulting corrections were smaller, rarely exceeding 5 meters in steep terrain and always 0 in more or less flat areas.Re *BEST* MESHWhen we started this FlightSimulator-Project using know-how from real-world projects, we have seen some SRTM scenery for FS with very bad interpolation of data voids, where certain authors were obviously lacking some mathematical background knowledge. Meanwhile, quality is increasing slowly. There are no completely unbiased criteria for what is the *best* scenery, of course. It depends on how you compare and what you're looking for. For example, in a few areas of large data voids (Himalaya) we used interpolation techniques that gives a certain "reduced safe envelope" forthe terrain, as it is required for more critical real-world problems. The resulting shape would look much nicer and more realistic by adding some generic "noise" and edges - but these would be pure fantasy and would increase the error compared to the real-world surface significantly. So we did not use this technique, though it would improve the "looks" and would certainly result in better reviews regarding certain high mountains... So the message is: we used considerable effort and avoided going "just for thelooks" to produce the to our knowledge best possible mesh for FlightSimulator out of the SRTM data. Major limitations come from FlightSimulator and SRTM data, but they affect all SRTM add-ons.I hope this sheds some more light and helps to understand the philosophy of FSGlobal 2005. Stefan Schaefer