very PJW-ish indeed
I almost though its q3 

...we are just bundling all the files together for the complete mission beta testing, and neg!ke has taken the time to set up a little profile over at ModDB. It has new shots of the levels (taken by him) that are far better than the rather spotty ones I posted not long ago.

http://mods.moddb.com/8904/travail/ 

i want to map like this... :\ i suck to hard... 

 

map 3 looks the best IMHO. 

Nuff said. 

I'm hoping that when I play the pak I will be so shocked from awe that I will transcend reality and live forever in the land of milk and polies.

If I don't, I'm going to demand my money back and sue for compensatory damages. 

...lol, I so needed one of those just now.

PS: We'll gladly give your money back, but fight to the death for one cent more! 

Finally stopped playing Stalker long enough to get this updated with various tweaks, changes, and bits of fixage. Added a new MH area to get it farther from RA. This is feeling pretty solid and is almost certainly the final beta, barring some major revelation.

The pk4 also includes a botfile for TinMan's Sabot v. 10 (extract to Quake4\sabot\maps\mp).

Screenies:
http://pjw.planetquake.gamespy.com/screenshots/pjw4tourney1ss1.jpg
http://pjw.planetquake.gamespy.com/screenshots/pjw4tourney1ss2.jpg
http://pjw.planetquake.gamespy.com/screenshots/pjw4tourney1ss3.jpg
http://pjw.planetquake.gamespy.com/screenshots/pjw4tourney1ss4.jpg
Bonus editor shot, if anyone cares: http://pjw.planetquake.gamespy.com/screenshots/pjw4tourney1editor.jpg

Download (map is ~13 meg):
http://www.fileplanet.com/dl.aspx?/planetquake/pjw/mapstorage/pjw4tourney1_beta2.zip

Non-FP mirror, courtesy of Scourge:
http://www.jwivie.com/q3wmaps/pjw/pjw4tourney1_beta2.zip 

I would like to put in a DM map for travail, I have the layout in my head, but I was just wondering whether or not its too late or still feasible to get anything submitted that isn't SP? 

Pics look excellent btw. 

...thank you for your kind offer.

As it stands, our three coders are just about finished pulling our progs together. We could say "go for it" but there's a good chance the mission will be ready to go before you can complete. Also, the DM maps that are included are there for specific reasons:

- qtdm1 answers the question "How much must one twist a new SP map (qtfin1) to make it playable in DM?"

- qtdm2_duel and qtdm2_ffa are natural extensions of brushwork stolen for use in qte2m2

- qtdm3 is an exercise in reskinning "qte1 style" (as well as some route expansion).

So, maybe we should look you up for the next project. 

Here's a shot from a work in progress, based of course on metlslime's design.

http://people.pwf.cam.ac.uk/~ajd70/fitz0001.jpg

The animations are barely started, and I'm not happy with the legs yet. Also this is a 'clean' version aside from the chainsaw, no blood splatters yet. One of the reasons for this is the skin is mirrored at the moment, and Rorschach style symmetric blots of blood might look quite odd. The skin has been designed so that the seams of the torso and head lie along the left edge of the skin. So if there's no way to make the ogre work as it is, then it can be mirrored and the left and right sides will line up exactly. It seems a bit odd to double the dimension of the skin just to accomodate the blood splatters, but never mind.

Criticisms are welcomed, hopefully with them it can be made into a faithful substitute model. 

Sorry to say that his face looks too "flat" IMHO... Maybe it is due to the shot angle... I don't know...
Anyway, it looks quite promising.. Keep it up ! 

That looks nice! The head is too human though in my opinion. The Quake ogre has almost no neck so the head is more near the body. At least that's how I think it looks.
I love the muscles! 

Ogres have no necks, but Fiends do have eyes ;-) 

the model looks quiet symetric while the original looks rather crippled.
For texturing I must say it suits good, knowing how hard it is to fit them.

But I would make it more damaged, somehowe a more reacktive attitude. 

you might want to add more shape to the main model. At the moment he still looks really low poly and simple, so there doesn't seem to be much point in changing the ogre for an updated one unless there is a significant improvement in both mesh and skin.

I'd concentrate on his characteristic parts most. Give him stubby legs and no neck with a chunky head and make his funny armour stand out more, then I'd probably do something with the chainsaw to make it look even meaner (than the original id chainsaw, which looks mean :). Try and think what cloud/carmack might have wanted him to look like had they had a bit more headroom. 

the ogre wore a blacksmiths apron rather than armour. 

I also imagined him with an apron sometimes (splattered with blood like a butcher's apron of course.) If there were ever a need for an alternate Ogre skin, that would be a very suitable look for him. 

Which is documented somewhere - chainsaw, leather apron, cannibal. 

I see it as a kind of leather 'apron' type thing for want of a better word, with darker brown cloth to the side, and also leather shoulderpads. But there is such thing as leather armour so I guess it serves as both. On the level of detail, the apron is at the moment contoured but not embossed, if you understand what I mean. There are polygons in the torso that correspond to it's outline, but they join directly to the polygons for the cloth.

The mesh is already at 940 odd polygons, and the limit is either 1000 or 1024, I forget which. Either way, there's not a whole lot more scope for detail, the mesh is pretty efficient, although some polys could maybe be shaved off the back, which isn't seen much, and bought to the front. The chainsaw is a difficult one, as I'm not sure how to do the saw blades on it. Going the zerstorer way of modelling spikes on seems out of the question, in a full model like this they'll just get mangled by the mdl format. So I guess it'll just have to be carefully skinned, but I'll keep thinking.

The single most overwhelming piece of feedback is that the ogre shouldn't have a neck, so I've acted on that one straight away. I agree that the face does look flat from the screenshot, but I'm gonna try and blame the angle and the quake model format for flattening it. Here's a 3/4 angle shot from the editor with a better view of the shape of it.
http://people.pwf.cam.ac.uk/~ajd70/ogregmax.jpg
If it keeps getting flattened like that when I export, I'll try exaggerating the features a bit so something shows up. A lot of this is going to be fighting against the twin limits of the mdl format, low polygon limit and poor vertex accuracy. Well, polygon limit is an engine limit, but still... 

That second shot looks great my man! Did you already get rid of the neck on it? Looks like it, but I'm not entirely sure. But anyhow, why such a high poly count for such a sad model format? Once you get into animating all that detail you've put into the face for example (omg a nose) will more or less get lost, as the vertexes will swim around on his face. Not entirely that bad ofcourse, but not far from it. I must admit Quake2 had close to this level of detail on its monsters, which has a model format suffering as much from wobbleitis as Quakes, so it might just work. Chillo makes models with around 1500 polys using the Quake model format, and they sometimes look just fine.

This post has no purpose, as I just contradicted myself, but whatever. Cool stuff man. 

Yeah, the neck is already gone, the model wasn't changed that much, I just gave it a slouch by adjusting the neck and head bones. I think I can get away with that level of detail better than a md2 could, because the quake 1 model format is in some ways less prone to wobble.

Both model formats do compressed coordinates, in that a single byte is uses to store each coordinate, from -127 to 127. The engine then applies a scale factor and a translation vector to these coordinates to give the model the correct scale. The difference is that mdl has a single scale factor and translation vector for the entire model, whereas md2 defines the translation vectors and scales per frame. This has advantages if your model grows and shinks - the problem is that models rarely do, they tend to occupy the same volume.

So md2s will generally try and maximise the accuracy per frame by setting the scale to the model's maximum extents/127. That sounds like a good idea, but if your model moves in the frame so that the outermost vertex in any direction changes, then the positions of every vertex get recalculated. Hence you get a lot of wobble. In mdl, if a vertex doesn't move, then it will snap to the same point on the grid next frame.


You could of course get around this in the md2, after all, it's a more flexable format in principle. For instance, you could just write the same scale/origin for all frames, emulating the mdl format. A better solution might be to write the same origin/scale for all frames in the same animation scene - or have an option that controls when you inherit the scale/location from the previous frame. That way, when the whole body is moving anyway, you can make use of the increased accuracy, and when it's a smaller motion of just part of the model you could hold things still. But I've not seen an md2 editor/exporter that offers these kinds of controls.

The trick for good results in mdl is either to keep something still, or make sure the animation is large if it moves. Then you tend to get good results. Exaggerating animations to a degree is usually a good idea anyway, so they are apparent on screen. If you need to do translational movement, then try and do it on the qc side of things, and constantly centre your mesh in the mdl. In any case, to make the mdls I export an md3 and covert it with quark. So I'm sure I can release the intermediate md3 as well, for people with engines that support 'em. Who knows, there might even be a normal/glossmapped version too... 

...since I am a complete donkey at doing anything for this community except playing Quake, since a little discussion on visuals of the Ogre arose and I did not feel like doing anything at work, I decided I'd do a SpeedSketching session and had a take on the Ogre. This is the result
http://shub-hub.quaddicted.com/files/images/the_silent_ogresketch.jpg
and this is likely to have me blown right out of the office, people looked at me weird while I was doing it...
I know it sucks bad, but, hey, it's only a wacom graphire! You can draw with more ease on your fingernails!
Anyyways, this is how, IMHO, da Ogre should look like, a little dumb or retarded (in a Friday 13th sense) while resting, and completely open jawed, all-out-of-it, when it's awake...
Hope someone out there enjoys my attempts at being fired...