Not going to lie, not really impressed with this route. As has been the trend recently, the hype leading up to the release is crushed and swept away by the performance and bugs. So far I've run the SD70ACe tutorial and the Something Old, Something New scenario, the scenario is where most of my criticism will come from below. I also haven't seen today's release stream, I literally fired this up after Thanksgiving dinner, so I don't know what's been said there. Physics Something isn't right with the physics of the SD70. Dynamics notch 8 down the 0.8% gradient feels as hard or harder than applying the emergency brake, with no air brake help. I think in 20 seconds I got from 40mph to needing to get the brakes off ASAP because I was under 10mph. On the return trip of the scenario, I threw it into notch 3 to get started, the rear of the train brakes was still around 74 PSI and I'm pretty sure I got up to 10-15 mph in less than 30 seconds. The first train also accelerated quickly, but it was downhill so I thought, maybe that helped...no. Performance I think at least half of my time running the scenario was spent under 30 fps. I actually noticed this a little last week on the preview stream with the PS5, which concerned me a little - now realized. I don't really mind 30-35 fps, but constant dips under 20, a couple as low as 5-10 isn't acceptable. Almost all of these dips were not stutters either, they were true frame rate drops that lasted several seconds. This is worse than the Dresden area. I keep HWiNFO up sometimes when running the game since the Dresden area thermal throttles my CPU, because why not, and there's no stress on any of my hardware in Sherman Hill. My technical guess coming from my programmer side is that here the scenery and its draw calls are so poorly optimized that my decent i7-9700 CPU stalls in communicating with my less than idea, but still ok 1050Ti GPU. Even so, the draw distance of this route is absolutely abysmal. I had draw distance setting on medium then bumped to Ultra and with the exterior camera I could see where the track drawing ends, that's the worst I've seen of any route to date. I guess draw distance had to be backed off because of performance, but Clinchfield has way more trees, and Dresden has a lot of assets, but they don't run nearly as bad. Something in the scenery assets needs some serious work. Bug The first signal you encounter in the scenario shows no aspect (sorry it's partially blocked by the wiper) Audio -There's no bell audio -I know this isn't using a real EMD 710-G3C engine sounds, that's fine. Upsetting, but there's nothing that can be done. But the mixing and implementation of what was used is really, really bad. Before I go into the issues, I will highlight a couple positives I'm glad were included. First, high idle tied to the reverser. Second, I like that the little pause was included between decreasing the throttle and the engine revving down, as is actually present in EMDs. Ok, now the bad. The acceleration curves are poorly implemented. It sounds faked, worse than anything in the game and in some cases the audio just jumps without much of a transition. While EMDs aren't quite as bass-y as GEs, there is still an bass-y undertone beneath the iconic turbo whine that is completely missing. (It's sort of there, but more like it was high-passed and only resonates in the higher frequencies). In addition, higher notches = higher RPM = louder engine. This specifically isn't new to the SD70, for some reason whoever mixes audio loves to turn down the volume on the higher notch audio in addition to stripping it of its bass, and it starts to lose it's character. Notch 6 at 50 mph, the turbo whine engine audio backs off like the throttle was dropped to notch 4 or so, but what is there of the engine "bass" remains as is. I got this to occur more than once, but only seems to happen if you're in notch 6.
I hope all of the problems & the bugs will be sorted out very soon. I've already purchase SH on Epic Store, at first i want to request a refund because the performance is the worst performed route/addons i've ever tried on TSW. But in the other hand, this addons has a great potential and fun to play.
As a temporary solution while the performance is being investigated, we have seen some players having more success with fps when their foliage settings are adjusted from ultra to medium.
This is why attention to detail is important. You're taking control of a very short train with 2 AC-locomotives. In notch 8 you have a maximum dynamic brake force of 2 x 106,000 lbf or 212,000 lbf. The train only weighs 1600 tons. That's nearly 7% of g (0.66 m/s^2 or 90 mph/min)! Incidentally 40 to 10 mph in 20 sec is also 90 mph/min or 7% of g. You would never drive the train in this way IRL. EDIT: Basically: some of the scenarios are a bit (to my knowledge) unrealistic in terms of the formations that are used. Usually they are overpowered. The timetable services are more realistic in this regard. Still, special trains are not impossible so the engineer has to adjust
1- I know you'd never actually do that, I honestly don't even remember why I ended up in notch 8, but I did... However, fair enough... I decided to take a minute and do some math myself, since you don't actually have 200000 lbf of force at 40 mph since it peaks much earlier (probably about half of that from what I've gathered on AC traction & dynamic brakes). So, I just integrated the changes in momentum and force over time from 40mph to 10mph, and it comes out to about 25 seconds. Note that is is done in steps of 5 mph with tractive effort rounded off for ease and linearly interpolated assuming peak at 20 mph), so give or take a few seconds for error. Combined with the 10-15 seconds for the dynamic brakes to actually wind up, and it's probably about 40 seconds total. I did, however, redo this scenario and it's not nearly as drastic as my OP, it's something very close to that 40 seconds (I think I lost track of the time and it seemed much quicker than it actually was), so that's my mistake. Cutting out the trailing loco it's about 1 min 15 sec, also about right. My understanding of physics comes from my meteorology degrees (same physics, different perspective) and that different perspective combined with the fact that DTG has only recently started getting physics more accurate (and as you said over-powered trains) throws me off a little
Also as an update, as I said above, I redid the scenario, and backed foliage down to 'Medium' (I only had it at 'High', so it wasn't that drastic of a change), and no, I don't touch my .ini file. I still encountered some areas of low FPS, but the drop gained between 7-10 additional FPS. Something about the grass assets needs some work. Clinchfield trees have significantly more foliage than this route, and I don't have this problem with the foliage setting on 'high', I may get drops into the 20s, but I don't mind that too much. Also, another update on the notch 6 engine sound thing, it seems it isn't tied to 50mph, I had it occur twice once at 54 and again at 64 (after dropping the throttle and pushing it back up). I thinking it's not tied to speed but how long you keep it in notch 6? I'd love to record this, but I'm not really set up to do any capturing while I'm gaming since I don't stream. Also, because of the poor draw distance, this bridge pops in very slowly and you get to see floating cars. I'm thinking some prioritization of assets is needed here so grass/landscape have a smaller draw distance than things like the track and bridges.
It seems that whatever this Grass is (looks kinda like wheat), whenever it's in high density, the frame rate plummets. When this grass clears or is sparse, I hold at a solid 60 fps, this one patch just outside of Cheyenne tanked me to 20-30. This is one of the more drastic areas, but whenever this is in high density I get between 30 - 45 fps max. Again just to reiterate my specs: i7-9700k, GTX 1050Ti, I've dialed foliage back to medium (usually have it on high with no issues), and no .ini customizations. Since I can't record a video, this is where the above screenshot is taken Staring at this patch of grass, FPS=30
Take a look at the root structure of prairie grass and you’ll see why digging it up created the dust bowl. That said your point on the prioritisation of assets is a good one. That slow pop up bridge is annoying.
Yes, bigger assets like bridges and big buildings could definitely load up from further away. Peformnace impact of them should be close to 0, considdring it's just a few of them once in a while.
I think that may help a lot, but foliage is very important for this route. I think wind also makes a huge difference, but I did not test that aspect yet. Turing off all engine.ini settings saved the game for me for the moment.
me Sherman Hill engine.ini settings [SystemSettings] r.ViewDistanceScale=1 r.SkeletalMeshLODBias=-5 r.StaticMeshLODDistanceScale=0.25 foliage.LODDistanceScale=1 r.MaterialQualityLevel=1 r.DepthOfFieldQuality=0 r.BloomQuality=0 r.MotionBlurQuality=0 r.HighQualityLightMaps=1 r.PostProcessAAQuality=6 r.MaxAnisotropy=16 r.TemporalAACatmullRom=1 r.TemporalAASharpness=1.0 r.TemporalAASamples=64 r.TemporalAAFilterSize=1.0 r.TemporalAAPauseCorrect=1 r.ToneMapper.Sharpen=1 r.SSR.Quality=2 foliage.DitheredLOD=0 Staring at this patch of grass, FPS=45
Is cab signaling working yet? It doesn't seem to be for me. I'll have a red/yellow cab aspect, which seems to limit speed to ~20mph, while passing three all green aspects with no divert or train anywhere close. I remember Matt quickly glossing over the fact that it was bugged in a live stream. I really hope they get this fixed. Also, has anyone noticed if the route is properly marked with speed reduction warning signs or track limit signs? I feel like ranting about a lack of a manual. DTG could do a way better job of managing player expectations if they would release manuals explaining the systems in each DLC. A lot of the negative Steam reviews included mention of not being able to move the train. An explanation of the new brake simulation in a manual would have eliminated most of those bad reviews. I have looked for UP cab signaling references and have been unsuccessful. That would be another manual subject for this DLC. I know the subject of manuals is listed in the roadmap, but that subject has been listed there for quite awhile. A good technical writer would solve this problem fairly quickly. Heck, pair that person up with a decent web developer to make an online manual. Right now, I think the TS version of Sherman Hill is superior and that is a shame.
No problem. What I don't know is if there are warning signals before the reductions or not. Because often I see other speed limit signs that my HUD doesn't show, so those could be a warning sign in advance. But I think those might also just be speed limit changes that only affect passenger/light loco runs. So for example 70/50 reduction to 60/50 or stuff like that. But even if there are no warning signs before the changes, there definitely are signs at the changes. Thankfully the route also has whistle boards everywhere they should as well, which is nice. ABout the in-cab signalling, I don't know what's wrong with it. I don't know how it works exactly, but never had a problem with it. Now I don't know what the TS version is like, as I've never played the route in TS, but I think this TSW version is very good. Not perfect, as seen on things like the SD40 or the completely missing scenery in the quarry or the framerate issues with the grass. But the majority of the route is really well done, and I think it's one of the best routes currently available for TSW. I always had problems with US freight routes, because I like them, but I will often get bored during the long trips. But this I am having a blast with, even during a 2 hour long run.
It's virtually impossible to compare the Sherman Hill route in TSW2 with the TS Classic version. Firstly, the TS version has the complete 3rd track. It also has myriad locomotives, both diesel and steam, like the GP 40-2, Ace, Big Boy and Challenger, which can be added on to the base package which has the ES44AC, SD 70M, SD 40-2 and the SW10 switcher as well as a lot of scenarios. It's a completely different experience to the TSW2 version. Of course, it doesn't have quite the graphics or the interactivity of the TSW2 version, just a lot more to do.
LLLLLLLLLLLLLLLLOOOOOOOOOOOOOOOOLLLLLLLLLLLLLLLLLL Leaves in winter...again. This is just sad at this point. And weird geometric patterns in the snow? Where's the mod that changed the trees to the correct ones for fall and winter?
Shouldn't the leaves disappear depending on the month you choose? So in one month you can have snow and leaves, and other months can have snow and no leaves? As for the ground geometry, I have also noticed this. How comes the ground is not rendered with edge smoothing on?
Add- ons for NA routes have been a rare event. In the early days of TSW2, there were extra locos for SFJ, LIRR and SPG. Correct me if I'm wrong, but I can't think of anything since September, 2020.
Well Peninsula Corridor is kind of done, not much to add there. Making new content for LIRR is pointless unless they fix the route first. Boston Sprinter and Sherman Hill are both quite new releases, hopefully it's a safe bet to say they will get new stuff eventually, because both have plenty of potential for other rolling stock. There is also Clinchfield, which could use some high-hood locos or something, but Clinchfield seems to be another one of those released and forgotten routes. No fixes for it, barely gets any mention at all.
Sherman Hill was my go to route in TS and over the YEARS there was plenty of add on DLC that came along to keep it fresh. So far, the TSW2 rendition of the route has been a blast to run as well. Only concern is over the FPS drop which appears to be with some of the foliage asserts. Would it be nice to have the third path, sure, but does it reduce my enjoyment of the scenarios and timetables of the routes they do provide? For me, no - having a blast finally getting a prototypical train length we are used to seeing in western freight railroading.
Thanks - I am redoing my modification I have to the .ini file just reducing foliage from ultra to medium improved FPS a lot with little reduction to the scenery
might have to put this route on the back burner until some patches comes cause i spent almost 3 hours doing a perfect run with what would’ve been a perfect score just to fail it at the very end cause the red signal was only FIVE FEET from the stopping point objective. absolutely ridiculous. who approved that? i can kinda understand if it was a passenger route but for American freight? yikes!
Understand the frustration, but can be avoid by managing your speed. Retired ship driver here and we had a saying we passed on to the junior officers learning ship handling - “Momentum is a B**ch”. Remember the speed limits in a yard are the maximum speed, not the target speed. You have a massive amount of weight behind you so even with a little speed, you have a massive amount of momentum. I try to creep up to the stop points with no more than 2 to 5 mph remembering, it takes a while for the pressure wave to travel the mile long brake pipe which means the braking force isn’t instantaneous. All that being said, a little more wiggle room would be appreciated
For this one I am going to ping TrainSim-Matt, because it's one ongoing complaint which is so easily fixed going forward. Please, Matt, put on your Exec Producer headdress and decree to the gameplay devs, carved on a bloody stone tablet, "THOU SHALT NOT place a stop marker bang on top of the signal!"
I don't know if this was a response to my comment or not since it wasn't quoted, but I was going very slow through the yard, but even when going only 5 mph like you said, stopping with a tiny window of 5 feet is going to be a challenge especially to new players or those who don't fully understand how American freight trains operate. Very glad to see TrainSim-Matt advocating for a fix for this though. Just saw he mentioned it in one of his streams he did. I really hope this is adjusted to at least 50-100 feet. 250 feet before a red signal would be preferred though.
Just have to make sure the rear of the train does not foul any switches, but with the length of tracks in the yards of SH, this shouldn’t be a problem. Another safety net is to save when you enter the siding…. That way if you over shoot the stop just reload and try again, shouldn’t have any problems with signaling and the save since you are all ready in the yard. Lesson learned the hard way in my early days
Sound advice from Leadcatcher, but if you SPAD and have forgotten to save, there's always the "return to last checkpoint" option (assuming you haven't disabled the checkpoint function), as there's usually (but not always) an achievement shortly before entering the siding.
I feel your pain. I noticed that also, stopped short and inched my way forward until I was on the stop marker, teetering on the edge of a SPAD.
Anyone else having issues with the reverser in the EMD SD70ACe? It keeps glitching and won't let me move it both w/ rail driver and with mouse/keyboard.
But "return to last checkpoint" is just an autosave variant of the exact same save routine and just as likely to be bugged.
We should have had the station type stop markers where if you stop short it doesnt matter and the game still reacts accordingly.
Yes I had that problem. My solution was to remove and re insert the reverser handle. That seemed to solve the problem.
I really like this route, but the grass is taking my FPS down into the 10-15 range, even with foliage setting on low and no ini mods. My PC specs are good; I get 25-30 in the yards with ultra settings. So, I am holding out investing a lot time here until that asset gets modified. Has anyone heard anything from DTG (or a modder) on whether this is being worked on and if a fix might come soon?
Yes, Matt has said that the team is taking another pass on the route to clear up performance hogs, especially the grass. What is not clear is if this means they are optimising the performance of the grass, or, in the approach we have seen in the past, simply thinning the vegetation. Cheers
