The questions answered in this issue are:
1. Is there a software tool for batch reduction models?
2. How to reduce the surface of the model with its own texture?
3. Synchronous baking of textures when the model is subtracted?
4. How much does the model subtraction affect the frame rate?
5. What is the true face of Tian Yuan Software (Enterprise Edition)?
6. What are the advantages of Tian Yuan Software (Enterprise Edition) compared with Online Personal Edition?
1. Preparation before batch reduction
1. Model classification.
There are a total of models for this batch processing33 pcsIn order to facilitate the later operation, we divide it into 30,000 (1 10,000 (24) (6 (2) total4 categoriesand save to four folders.
2. Observe the model and confirm the selected functions.
In order to determine which feature to use to process the batch of models, we need to first observe and analyze the models. In Maya, we can see that the wiring of these character models is relatively dense and detailed, and these details are created through the model, and there are also textures. Therefore, for the subsequent processing, we will choose the "smart PBR" function in Tianyuan software, which can be directOutputs optimized results with maps
2. Batch reduction model
1. Install the batch processing software - Tian Yuan Software (Enterprise Edition).
At present, there are few software on the market that can reduce the surface of batch 3D models, and the performance of Tianyuan Software (Enterprise Edition) is relatively superior. It canInstall independently to your own server, so you canWork with models offlineto protect your sensitive data.
2. Optimize the model in batches.
As mentioned above, we use this treatment"Smart PBR".Mode. Just now we have divided the 33 models into 4 categories according to the model size, and we can batch process them in 4 times. The steps are the same and very simple:
First of all, click the "Smart PBR" function, then directly drag and drop the models of the same category to the workspace, set the parameters according to the requirements of the reduced surface scale, select the way to expand the UV according to the characteristics of the model, select the format of the model to save according to the subsequent requirements, and finally click "Next". After completing the above steps, you just need to wait quietly for Tianyuan software to process it. We can also see the progress of each model along the way. In practice, this33 pcsIt probably cost before and after the modelMore than 10 minutesOn average, it cost about each oneTwenty or thirty secondsThe work efficiency is very high.
3. Sort out the model after the reduction
1. Decompress the model.
Click on the folder where the optimized model is located and extract all the archives. We clicked on each folder and saw that there was still one in the optimized model5 stickers, including:Normal, Ao, Color, Depth, and Transparency maps
2. Delete unused content.
This batch of models is only needed hereNormal maps and AO mapsThat's it, so delete the other textures and the archive.
Tip]: When deleting related stickers, you only need to copy "color, opacity, thickness" to the search bar of the previous folder, and then you can quickly delete the searched stickers.
Fourth, the follow-up processing of the model
Import the model into UE4, compare the original model to make shaders, and observe the differences between the model and the original model to see if the model meets the optimization requirements. And this batch of models processed by Tianyuan software has a very good optimization effectThe details have been preservedDown,The wiring is also very evento place it in the scene and the original modelNot much difference。As a result, all models can be replaced with the original model.
5. The difference between the number of frames before and after optimization
1. The frame rate of the original model.
Hide the other content, keep only the original model, turn on "Show FPS", turn on full screen, and see that the frame rate of the 33 metamodels is approximatelyframes
2. The frame rate of the optimized model.
In the same step, you will see that the frame rate of the optimized model is approximatelyframes
The frame rate is significantly higher than before optimization. Don't think that only a few frames are optimized, but if you optimize a few frames in every part of the scene, then this project will runThe cost of performance can be greatly reducedFinish. So, the optimization of these models is for the whole projectIt is very necessaryTarget.
6. Advantages of Tian Yuan Software (Enterprise Edition).
1. The efficiency is very high.
Tian Yuan software can be used for the modelBatch processingAnd the processing speed is also very fast, which greatly improves the work efficiency, so it is very friendly to individuals or enterprises who need to complete tasks in a short period of time; In the long run, automated processing replaces manual processing, reducing the labor cost of enterprises.
2. Privacy is very good.
Tian Yuan Software can be carried outPrivatized deploymentTherefore, the model can be processed offline, which is very important for enterprises with strict data confidentiality requirements, especially large enterprises and state-owned enterprises.
3. Powerful.
The model to be treatedNo restrictionsRegardless of the size of the model, the number of faces, or the number of sub-objects, Tianyuan software can handle it, and the function is very powerful.
4. Efficient office.
Tian Yuan software can also be carried outMultiple clients are used concurrentlyMembers can log in to the processing model together, making the office more efficient.
5. Simple operation.
The operation interface of Tianyuan Software is very concise and clear, and the entire operation can be completed in 3 simple steps, and the completion progress of each task can be seen without training, so that enterprise members can quickly get started.