• Posts

  • Joined

  • Last visited

  • Days Won


franekma last won the day on July 29

franekma had the most liked content!

About franekma

  • Birthday 01/23/1993

Recent Profile Visitors

449 profile views

franekma's Achievements

Private 1st Class

Private 1st Class (2/11)



  1. Making a minimap and a tactical map for Gaia Canyon has been an ordeal. Initially, I simply followed the instructions in the TFE Editor documentation. Sadly, during the playtest (thanks to all who took part, by the way - I appreciate it!) it became clear that Gaia Canyon's complex layout meant that the minimap needed to give players who hadn't played the map before (both veterans and new TFE players) more help to navigate the level. So, I set to work and created the improved minimap for the [2021 - 06 - 29] version. Take a look at the screenshot below: Sadly, even though the resolution for this minimap was the medium-detailed version, 2048x2048, it still wasn't terribly clear for the player. For starters, the routes shown by arrows and dots weren't always contrasted enough with the greenish landscape. It also wasn't very helpful in showing the player all the places in Gaia Canyon that they could go. Then, I made this minimap: Oh, this one was beautiful. The routes were clearly defined - blue routes had arrows showing the directions blue players needed to take in order to get to the centre of the map; green routes had arrows showing the directions players needed to take in order to get to the outposts. Smaller white arrows were used to show where players needed to jump. Overall, I thought this version was fantastic and looked the most professional. Sadly, I had two problems. First of all, my graphics card, the GTX 980 (4GB VRAM), ran out of video memory while I was 'capturing' - or generating - the minimap. That's why the landscape lacked detail in 3/4 of the minimap (compare the detail of the landscape of screenshot 1 and 2). Actually, I preferred the way the landscape looked with fewer details, so that problem seemed to solve itself. Sure, the one remaining quarter of the minimap would be detailed and look strange, but I figured I could live with that. However, the second problem was too big an issue to ignore. For this minimap, I had chosen the maximum suggested resolution from the TFE Editor documentation: 4096x4096. And oh boy do those circles look circular! For all the detail and beauty 4096x4096 afforded, it came at a great cost. The file size for the map grew from 302MB to 792MB! So, I went back to the drawing board: This time, I knew adding arrows and dots for routes wouldn't work. I needed to use the minimum resolution to keep the file size as low as possible. I also wanted the minimap to be as clear as possible. Instead of making dots and arrows which wouldn't really resemble dots and arrows in-game, I decided to black out the areas which were not navigable. I also hid the landscape and replaced it with a flat, light grey texture; any area onto which you must jump or can only jump over (i.e. jumping down a small hill) was dark grey. Any elevated areas (inside towers and on the bridges) were white. I also went through the painstaking effort of retexturing every object in the game and replacing it with the dark grey texture, just to add clarity. So, the final version of the map is 180MB with a very readable minimap. I look forward to hearing feedback 🙂
  2. I'm not entirely sure what you mean, Anthony EX. Can you elaborate?
  3. I'm playing video games tonight, but I'll be up for doing it tomorrow! 🙂
  4. I agree. It took trial and error. Would you like me to make that PDF?
  5. Very good question. These steps have modified collision parameters; there is no collision for players, but there should be for projectiles. This means that players can walk through them and their screen does not jitter or bob up and down, but bullets should still hit them. I did this by creating a custom asset because Psycho and I weren't sure whether the Eddie-modified parameters of a TFE asset would be overwritten by the core game. If you'd like help on how to achieve this effect and a few examples elsewhere in Gaia Canyon, I can prepare this for you in a PDF.
  6. Gaia Canyon is a fast-paced flanking map, a re-imagination of one of my old creations from the BHD days. I have been very sparing with decorations and have tried to optimise navigation around the level as much as possible. For server admins: Be sure to download the latest version. Check the changelog for information. The latest version is GaiaCanyon_20210726. For level designers: Download the editable .uproject files and assets. Feel free to modify and republish the level to suit your needs. Download here! It has been designed for the following game modes: TDM CTF TKOTH TC To do: Changelog: I have added some screenshots below: ^ Gaia Canyon Loading Screen ^ Gaia Canyon Approach to Base ^ Gaia Canyon Outposts Alpha and Delta ^ Gaia Canyon Connector between Outposts and Mid ^ Gaia Canyon Outposts Bravo and Charlie ^ Gaia Canyon Mid ^ Gaia Canyon View of Base CTF ^ Gaia Canyon View of Base TDM, TKOTH, TC ^ Gaia Canyon From the outside above Connector ^ Gaia Canyon From a distance
  7. franekma

    BetaTest Session

    I'll be there. I'll need a good few hours before the match to download the update.
  8. I recall something along these lines being in the Titanfall Attrition game mode, in which players engaged in all the rooty tooty big boy shooty for 15 minutes then tried to escape into the exfil VTOL. If the exfil element attached to TKOTH/KOTH defines who wins the game, then it reduces the worth of the TKOTH/KOTH part of the match. Why should I fight for the zone, if I can just get on the helo? If the exfil element attached to TKOTH/KOTH does not define who wins the game, then players won't be motivated to try and get on the helo. Why should I risk getting on the helo if it doesn't matter anyway? If the exfil element is worth x points, while the TKOTH/KOTH part is worth 5x points (5 times as much), then it would make some sense. Perhaps exfil could serve as the decider in a tie break between two teams. However, if I'm really good at TKOTH/KOTH and I fought very hard for the zone, why should my victory be determined by someone being able to shoot a helo from the sky, a skill that's only useful once per match. It's kind of like Harry Potter Quidditch - where catching the Golden Snitch is worth more than anything else in the game, rendering the other scoring system completely pointless. The losing team in the final moments of regular play could just plant one of their players by the helo and wait for the other team or just destroy it before the other team gets on. With those arguments out of the way, there could be a whole different game mode (ignoring TKOTH/KOTH): One team needs to exfil; the hunting team needs to stop them. The exfil team can choose between an APC or a helo. APC is armoured and can withstand 3-4 RPGs but is slower and has only one defence turret. The helo is not armoured and can only withstand 1-2 RPGs, but is harder to hit. The exfil team have to choose between either the APC or the helo and have to fight their way through the map to reach them. Both the helo and the APC can be defended by the hunting team, but they cannot be destroyed by the defending team until the vehicles have been activated by the exfil team. Both teams respawn like in normal games and even have PSPs. Scoring is based over two games on the same map. Whichever team exfilled the most players wins. If it's a tie, the tie break may be based around points or map control or something. I personally think my idea above is garbage, and realised this when I was writing it, but I figured I may as well send it anyway. Maybe it will help someone come up with a better idea. Overall, I'm happy with Accumulative TKOTH and I would still like to try RJS' Attack and Defend.
  9. What you are warning against is exactly the kind of shenanigans I used to get up to when I was playing with close friends in the server. It was always a laugh, but I never did it to random players. The reason for a push mechanic beyond what I suggested previously is also to prevent too many players clipping into each other.
  10. There is no reason for collision boxes of a prone player to prevent another from jumping over him. I imagine this is a bug and will be fixed at some point. A solution to the collision problem in the tunnel map - make the player-player collision boxes much smaller (imagine a small 5cm diameter cylinder running through the height of the player's model) with a gentle push function when you get too close. This would mean that you can brush past players who are blocking the way, but you also can affect their movement/position. DUZ will be able to correct me, but I think BHD had this kind of function. I don't ever recall getting stuck, but I could still knock players off of rooftops and other high places. Yeah, that was a lot of fun 😛
  11. You could always implement various classes or categories for the badges based on quality. For example, gameplay-related achievements could all be coloured gold/red, while less important badges could be black-and-white or a more subdued colour. Further, the badges could be ordered automatically according to their importance, gameplay-related badges being the most important. This means that when you browse someone's profile, you could see at a glance what kind of badges they have and, while hovering over them with the mouse, you can see a little more information. I definitely agree with Lost that obscure badges will just get in the way if they aren't organised well. Either way, top idea. I love it.
  12. I'm flexible, so I can make time whenever, usually. There are a few hours per week (Wednesday 19:00-20:00 CET; Thursday 15:00-18:00 CET; Saturday 09:00-12:00 CET) that I cannot reschedule.
  13. I didn't mention - the only concern that I have with lllama's idea here is that the gap in performance between 1st place and 2nd place from one game to another will be different. The gap between 1st and 2nd and that between 2nd and 3rd will be different. I don't think this statistic would be relevant in the competitive realm; I cannot imagine comparing professional CS players on their scoreboard position at the end of a game. Of course, TFE is a completely different kettle of fish. Who knows, maybe it could apply. Either way, it's a good starting point and a way of approaching player stats objectively - what lllama set out to do.
  14. That's an interesting idea, llama. I think there will be a point at which a line will be drawn with regards trying to compare all players across all game modes. As discussed, the idea of a Skill rating, or some arbitrary number which suggests how good a player is in any given game mode will likely be a contentious issue among the casual players and be completely disregarded by competitive players, since it just isn't accurate enough. That said, I personally am still not a fan of using Babstats' approach of cumulative data without taking into consideration play time. It's nice to have all-time leaderboards, but this isn't a perfect system by which player skill/efficacy/input is determined. Finally, speaking of what I personally find interesting - introspective stats. I love the idea of being able to browse my statistics and compare them to themselves. These would be questions like: Which weapons am I most effective with? What are my best maps? What time of day do I play at my best? How do I compare this month versus last month? This one might be a bit farfetched, but: Using heatmaps, where on a given map am I most effective? Where could I change my play to help my teammates more effectively? Oh boy would these kinds of stats be fun!
  15. There are some great points here. I’ll start with Lost’s comments about the value of certain stats versus others in the non-competitive realm. The discussion before had been about how you can match two players’ stats across multiple game modes. I argued that this was incredibly difficult, and I haven’t seen it done very well. If I’m not mistaken, Lost has demonstrated that it seems more viable to separate these leaderboards and, using total play time either of the game or of a certain game mode, to interpret the cumulative data, such as kills, deaths, assists and zone time. The only issue that I take up with this approach is that if these stats are separate figures, not everyone will consider the total play time. Your average Joe is more likely to make sense of one number when comparing it to his friend. The problem with this is, however, that it can be misleading, as I demonstrated with Battlelog. Skill or Score per Minute doesn’t paint the whole picture, but it is a way to compare players at a very basic level. If players are interested – and we’ve got to respect that not all players are – they can then go and explore the rest of the stats. My suggestion is that an easy-to-interpret statistic will go a long way to get users interested in observing their statistics and comparing themselves to other players. This could be a statistic that incorporates total play time as a means to interpret cumulative data as discussed above. Ultimately, it needs to be something that’s absolutely relevant to TFE and TFE’s game modes. I don't have the answer here; I'm pooling ideas to help someone smarter than me find it! I’ll just highlight that my exploration of Battlelog was intended as a way for casual players to observe their statistics. I don’t have much experience with esports, so your understanding of HLTV and CS is very valuable. These are all statistics that are used for a single game mode in CS, so it’s imperative that whatever method RJS uses is tailored to TFE. We’ve got to wait until the game mode is finalised before we can decide for certain what is useful. As we can see, the standards by which players are judged in CS have changed considerably in the 20-odd years the game has been around! DUZ suggested making a formula which balances all the game modes together – yes, it’s definitely the most complex aspect of the design. Off the top of my head, I can imagine having humans interpret vast amounts of data to decide where the line of comparison is between the different game modes. With vast amounts of player data, you will be able to see trends and patterns. Using percentiles, you could map out a way to compare scores from different game modes (i.e. the 5th percentile achieves an average score of, say, 700 on TDM, but 950 on TKOTH). This is obviously a very crude way of looking at it; the mathematicians and statisticians among us could work out a better way. I’m just a linguist! DUZ made a point that Battlelog was very confusing. Yes. Yes it was. For my post, I originally recorded a short video and made a GIF of me navigating it, but I quickly realised that it might confuse anyone who has had the good fortune not to browse the site. I tried to present the information to you without all the unnecessary jumping around and navigating. I’ll be clear: I hated Battlelog until about a year ago, when I started playing on a friendly casual server with players who were fairly competitive and got me interested. I persevered and now feel fairly comfortable navigating Battlelog. RJS, please do not copy the UI of Battlelog. It’s bad. SPM and KPM are very broad and do not do much to explain how good a player is on a given game mode. Take Rush, for example. In Rush, the attacking team needs to destroy two objectives in order progress on to the next two objectives. We typically played both sides on our server, so the team with the most objectives defended or destroyed over the two rounds won. We would judge our performances in real-time on TeamSpeak, and something as simple as a public player leaving after the first map and not playing the second map was immediately taken into consideration as making a clear-cut result difficult. None of this is available in Battlelog. Furthermore, you get 400 points for planting an explosive on the objective. You get 100 points for making a kill (extra points for spotting players, extra points for assists, suppression assists and so on), meaning that it’s easy to be the player with the highest score on the leaderboard while not actually going for the objective. DUZ’s screenshot of DF1 portrays this problem very accurately. When I played CS with a fair degree of skill (2015-16), I was an entryfragger, so I usually had the most deaths in the game; the information I gave on voice chat did a lot to help my team win rounds. This information was not only unavailable, but it’s physically impossible to calculate the quality of voice comms in-game. Yet, in the professional CS scene, you wouldn’t compare an entryfragger to a support directly, because there isn’t an accurate apples to apples comparison. It’s as though these players are playing different games. Their objectives are substantially different. Again, as before, once we know what the finalised game modes are, we’ll have a better idea of how to understand and interpret player data. As I mentioned before, we’ll also likely need a lot of player data in order to find out what we’re looking for. The Skill rating is troublesome. I never did look up exactly how it was determined, but I was certain that your performance over the last few games was taken into account. In my previous post, I wanted to imply that it wasn’t simply a win or loss that influenced this, but rather your performance in the game. I just made a quick generalisation that because he lost, he might have been playing badly, or worse than he would normally. Of course, this is not always the case and was an off-the-cuff judgement. However, DUZ’s comment doesn’t seem completely accurate to me, the comment about the exact nature of the Skill rating regarding the last 200 matches and that the last few games aren’t likely to influence the Skill rating. In my time playing BF4, even after I’d played over 400 hours, I’d notice that my Skill rating could jump up and down by over 50 points in a single evening, across just a few maps. This is significant enough to drop down or rise up in Division rank, so I do believe the latest games do bear a significant weighting. However, my data is anecdotal – I never cared enough about my stats in BF4 when I still played it to analyse this. The reddit link that DUZ posted seems to confirm my suspicions: “since Skill is calculated on a running basis from each game, you can have much better overall stats than someone, but still have a lower skill. It is very much a "What have you done for me lately" stat”. Also, I completely agree with you, DUZ, about players doing ‘the dirty job’ and the danger of being considered selfish or a bad teammate due to their role. There is an interesting point made by Thooorin, a well-known journalist for esports, who argues that numbers on a screen are not the best way to judge the value and skill of a player. They shouldn’t be used as a definitive answer to the comparison between two players, but rather an introduction to a discussion. The true value would be uncovered as a result of reviewing demos (in-game audiovisual footage of matches played) and passing judgement on input made in a real match. To demonstrate this point in terms of TKOTH – a hypothetical player playing a hypothetical match on Eye of the Dead could spend the entire game occupying the enemy’s PSP and harassing the enemy team in their base, preventing them from even getting to the objective. Looking at the figures, they didn’t make a single attacking or defending kill, nor did they make up any zone time. All they did was get kills (and give information to their team and help prevent the enemy team from even getting to the objective). As such, in the competitive scene, I don't think there is a need for a single number by which we judge all players, regardless of game mode and role. Finally, Lost’s idea to show results and stats for all official clan matches in a single place would be good. That’s something I’d like to see, for sure. If anyone would like me to find that Thooorin video, let me know and I’ll look it up for you. It could be quite informative for this discussion.