Jump to content

HDT HH Performance Patch

Sign in to follow this  
Lupine00

1,329 views

This is a trivial little hack that stops HDT HH overloading your Papyrus with its crazy 0.3 second interval updates.

You might see a bit more lag on heel height updates, but overall I don't notice any difference.

hdtHighHeel-performance-patch.7z

Overwrite files in your existing HDT HH to install.

 

Sign in to follow this  


15 Comments


Recommended Comments

That's cool!

Though, whenever I install a mod, I just convert the HDT HH high heels to NIO HH instead, which is a part of the Racemenu and doesn't stress your game at all (subjective opinion).

Share this comment


Link to comment
23 minutes ago, Nagib said:

That's cool!

Though, whenever I install a mod, I just convert the HDT HH high heels to NIO HH instead, which is a part of the Racemenu and doesn't stress your game at all (subjective opinion).

How do you do that, please?

Share this comment


Link to comment
5 hours ago, mimichacha said:

HDT HH is outdated years ago, what's the point of using it?

Isn't it the case that, unlike the NIO HH implementation, HDT HH doesn't cause the character to float when lying down or kneeling?  Which is partly why it's so much heavier?  Or has this changed?

Share this comment


Link to comment
5 hours ago, legraf said:

Isn't it the case that, unlike the NIO HH implementation, HDT HH doesn't cause the character to float when lying down or kneeling?  Which is partly why it's so much heavier?  Or has this changed?

I highly doubt that, especially for kneeling without floating, which would require a much more sophisticated system to lift each foot individually. However, if I remember correctly, neither NIO nor HDT can do that. Both of them simply lift the whole actor.

Share this comment


Link to comment
2 hours ago, mimichacha said:

I highly doubt that, especially for kneeling without floating, which would require a much more sophisticated system to lift each foot individually. However, if I remember correctly, neither NIO nor HDT can do that. Both of them simply lift the whole actor.

Not that sophisticated - but it tries to gauge the angle of the lower leg (based on "calf bone"), and calculate whole-actor height offset based on that angle, which is how it could give no "lift" when the lower leg was horizontal - no idea how it handles both legs, take the average?  At least, I remember a discussion of this long ago, as part of the explanation for HDT HH's potential advantage.  There is a "fix" for NIO heels which removes the lift when sitting or swimming, but that doesn't help in other situations - crawling, sleeping in heels, etc.

Share this comment


Link to comment

thx for the patch!

I still have to have HDT high heels in LO for some mods (or many atm lol) from time to time

Share this comment


Link to comment
6 hours ago, legraf said:

Not that sophisticated - but it tries to gauge the angle of the lower leg (based on "calf bone"), and calculate whole-actor height offset based on that angle, which is how it could give no "lift" when the lower leg was horizontal - no idea how it handles both legs, take the average?  At least, I remember a discussion of this long ago, as part of the explanation for HDT HH's potential advantage.  There is a "fix" for NIO heels which removes the lift when sitting or swimming, but that doesn't help in other situations - crawling, sleeping in heels, etc.

This potential advantage of HDT HH is but a theory, and it was never implemented, right?

 

Share this comment


Link to comment

The reason I tolerate HDT HH is because it's just too much bother to convert an entire pack like NiNi to NiOverride.

And if it's updated, you may have to check models and do it again. Too tedious to bother.

 

For one off models, like DeserterX, I do the conversion.

 

For whatever reason, a lot of high-end armor modelers seem to cling to HDT HH. Maybe they are under some impression that it's better, or maybe they simply find it easy to understand. The big advantage of NiOverride is that the modifications are only in the NIF, so any ESP that uses that NIF immediately gets the benefit.

 

2 hours ago, mimichacha said:

This potential advantage of HDT HH is but a theory, and it was never implemented, right?

No. It's definitely implemented, and it's why the updates were set to run so fast - so that this calculation can be adjusted responsively - but as it relies on Papyrus, it doesn't really work as intended. The results with my increased update delay are barely any different in terms of response in most cases.

 

I rarely even notice the main offset delay when you put the boots on unless I'm thinking about it.

 

In any case, the original high update rate isn't needed, justified, or effective.

Share this comment


Link to comment
On 2/27/2020 at 7:25 PM, Lupine00 said:

 

No. It's definitely implemented, and it's why the updates were set to run so fast - so that this calculation can be adjusted responsively - but as it relies on Papyrus, it doesn't really work as intended. The results with my increased update delay are barely any different in terms of response in most cases.

 

I rarely even notice the main offset delay when you put the boots on unless I'm thinking about it.

 

In any case, the original high update rate isn't needed, justified, or effective.

 

That's a big news to me. I should give HDT HH a try when I have the chance.

Share this comment


Link to comment
7 hours ago, mimichacha said:

I should give HDT HH a try when I have the chance.

Unless you have a massive armor pack that uses it (Nini being the most obvious) then maybe try to avoid it?

 

There are reports of this 'hack' making a big difference to script performance, which means that HDT in its stock form is relatively toxic - which I'd always suspected.

It's had a poor reputation for a long time.

 

I wouldn't ever choose it just because it's better, because the drawbacks are considerable to get a small visual improvement.

Share this comment


Link to comment
On 2/27/2020 at 6:25 AM, Lupine00 said:

The reason I tolerate HDT HH is because it's just too much bother to convert an entire pack like NiNi to NiOverride.

And if it's updated, you may have to check models and do it again. Too tedious to bother.

 

For one off models, like DeserterX, I do the conversion.

 

For whatever reason, a lot of high-end armor modelers seem to cling to HDT HH. Maybe they are under some impression that it's better, or maybe they simply find it easy to understand. The big advantage of NiOverride is that the modifications are only in the NIF, so any ESP that uses that NIF immediately gets the benefit.

 

No. It's definitely implemented, and it's why the updates were set to run so fast - so that this calculation can be adjusted responsively - but as it relies on Papyrus, it doesn't really work as intended. The results with my increased update delay are barely any different in terms of response in most cases.

 

I rarely even notice the main offset delay when you put the boots on unless I'm thinking about it.

 

In any case, the original high update rate isn't needed, justified, or effective.

The problem has always been that you'd get a CDT on decapitation with NIOverride. I'm mainly playing FO4 now so there's no way for me to verify it, but I know it was happening on the most recent time I've tried to switch over from HDT.

 

There were mentions of it being fixed as well, but it seems like I'm not the only one that still gets CTDs from it.

Share this comment


Link to comment
1 hour ago, BlindingBlizzard said:

The problem has always been that you'd get a CDT on decapitation with NIOverride. I'm mainly playing FO4 now so there's no way for me to verify it, but I know it was happening on the most recent time I've tried to switch over from HDT.

 

There were mentions of it being fixed as well, but it seems like I'm not the only one that still gets CTDs from it.

With my update hack, HDT HH causes almost no impact.

At least, I notice no perceptible impact from it in play.

 

With HDT HH in its standard form, I had significant delays between script launch and script completion - script lag - and I also had problems with Story Manager processing working properly. This manifested directly as acute failures in Maria Eden and other performance sensitive mods.

 

I confirmed quite simply that this was happening with HDT HH installed, and when it was stopped from updating, the problems went away. It was a very obvious difference.

 

Simply updating at a more pedestrian rate makes a surprising difference, and the lag in HDT HH updates is not particularly noticeable.

 

But I also use NiOverride.

It's not like I'm going to convert all the newer armors to HDT HH, which is one reason why the CTD issue isn't a valid reason to promote HDT HH; the ship has already sailed.

We need to cope with both, or do a lot of armor conversion.

 

Personally, I don't have issues with the CTD tho. Maybe it's because I simply don't take the decapitation perk?

 

Looking at the C++, I can't really see why NiOverride would have an issue that HDT HH doesn't, as the underlying code that modifies the skeleton transforms is extremely similar in both mods. I guess it must lie in the NIF structure changes required to add the NiOverride data (or rather, some issue in parsing that).

Share this comment


Link to comment

Press X to doubt.

 

Grandpa may still love his old brick phone but it doesn't mean that we have to. Besides it isn't that hard to fix the issue by converting it to NiOverride. Just keep a record of all the heel heights and change it as necessary. Now if only Bodyslide would stay consistent. All the outfits are clipping though my character's neck which never happened before. Technology is a royal pain in the ass. Also I haven't been playing Skyrim for some time. No doubt I'll get back to it soon. Stupid Todd and your stupid sexy mountains!😩

 

Anyway don't mind me. keep doing what you're doing.

Share this comment


Link to comment
×
×
  • Create New...