cloud capacity and price.

[super gnu]

This might already have been answered, strange if not, but i couldnt find an answer so im asking.

I appreciate we can test a few frames etc to get an idea of rendering costs for animation.... but overall time to complete job? It seems I have no way of knowing how many machines will be on my job, and therefore how many frames will be rendered simultaneously.. i also have no way of knowing if your capacity is maxed out on some other big jobs..

It might be the case that you have "infinite" capacity, scaling as needed, and you will always render all my frames at the same time?


I don't see this clarified anywhere. Its rather important to be able to tell how long a job will take, at least roughly.


also, i assume you have done some studies on how your prices compare to the various renderfarms out there.... are you generally equivalent? cheaper? more expensive due to convenience?

It seems a very useful system, and I would consider switching to it from my current farm provider, however im rather concerned the layer of "incompatibilities" which going via vraystandalone brings to the table.. ive already had to partially rebuild my first test scene due to incompatible maps... at least with a vray/max farm, you know if your scene works at home, itll work there.

[super gnu]

ok the capacity thing i believe is (partially) answered in the FAQ:

How many frames can I render simultaneously?


You can render up to 100 frames at the same time.


so. does this mean if i submit up to 100 frames, it will render them all simultaneously, every time? and if i render 200 frames, it will render 100 then 100?

[vladie]

Hi super gnu,

Thanks for the questions and reaching out to us!

With respect to your question about the overall time to complete an animation job, there are a couple of aspects as I see them:

1. The average time/cost to render one frame. For that we have this guide which we advise users to use. Besides time/cost it gives idea whether there might be any discrepancies between the actual Chaos Cloud output and the users' expectations.
2. The total time of completing an animation job. It includes the render time of one frame but also the waiting time between the different frames.
   I see you found in the FAQ the current limit for simultaneously rendering frames/machines in Chaos Cloud (one frame is always rendered on its own dedicated machine). And this limit is valid every time. As this is a parameter, we can customize it and when clients ask us we decide together what might be the best parameter for them.
   It is more or less a kind of a safeguard for the users. The reason is that the render machines can pick up and render really quickly, and if there is an issue in the animation (e.g. visible in the Live preview in Chaos Cloud), though the users will be able to pause/stop the rendering, a number of frames will render while the process is stopped. Of course if you render just a few frames first (as described in the guide), you will be able to notice such issue before spending many credits. Then you will be more confident to use a bigger limit for simultaneously rendering machines. There are such Chaos Cloud clients - for some we have increased the limit up to 300-500 simultaneously rendering machines.

In regards of the price, it's difficult and not entirely true to directly compare us with the other farms as we have different features and workflow. As you have noticed some features, which are supported in the render farms, are not supported in Chaos Cloud and vice versa. Also please have in mind that by buying larger packages of credits, you get bigger discount, hence the cost decreases. So its best to use the Chaos Cloud trial credits and the already mentioned guide to calculate what will be the cost in Chaos Cloud, whether we support the needed features and the output is rendered as expected. We also have similar guide for still scenes.

Lastly about the incompatibilities, you can always reach out to our Support team and they will assist you by either finding a reasonable workaround or logging an issue in our projects. Sometimes it's a bug or a missing feature which we fix.

Best regards,
Vladie