Pricing competitiveness TF vs AWS and others

[sensorium]

I wanted to see how TF compares to other known brands of hosting providers. I’ve included the 60% discount on TF although it’s not ideal(any larger deployment will have difficulty keeping 3 years of TFT at hand, so ideally there could be a size bonus for contracts?), and others have also discounts. E.g. Amazon has a good discount if prepaid for a year.

But all in all (assuming I got the numbers right) does not look especially exciting for customers.

Originally I included a table, but it didn’t work in bbcode format so image:

[kristof]

please check on the following calculator

thank you for the feedback, we are playing around with some nr’s for our next alpha 5 release which goes on testnet this weekend or next week.

the nr’s in the link are not final but give idea, all feedback welcome

we will do a GEP for it (Grid Enhancement Proposal).

[kristof]

and with 60% discount

would be nice if you can let us know if the pricing is competitive like this, many regards

Kristof

[sensorium]

Still looks bad using that calculator(only the per 1TB storage price looks okay):

In most cases using a 40% discount it’s not enough to be competitive with classic hosting and they don’t ask for a year long contract for that price (Amazon gives also good discounts if prepaid 1-3 years). I know you want to maximise the holders but it’s killing the reason why someone would want to use it.

I’m not an expert, but from what I can see I think the root cause might be that you are manually setting the price of TFT. That might sit well with miners(in the short term), but in reality it will hurt the project. It’s just a matter of time when you are gonna run out of money. I know, it’s nice to sell those minted TFT’s for 0.80$(current price) but if you’re not in for the quick buck but want to change the course of the project here is my suggestion:
Remove the 50TFT barrier on the bridge to BNB, set it to 1TFT like it’s on the inbound(there is no real world reason why it needs to be 50 except to minimise miners cashing out), make it instant and more people will cash in their TFT. In turn lowering the TFT price and making these numbers look better. If there is a real supply-demand things might actually work. Yes your TFT’s will be worth less and mining won’t be so profitable but in the long run things should work out once there is good adoption. And right now I see everyone just cheering farming and barely anyone mentioning actually using it, and without using it it’s just looking bad.

[rad]

Hi Guys,

I wanted to add my example. I have few Helium miners, one of the issue people have is need of public IP. Not all ISP are having this. Personally I’m using VPS from Ionos for 1,23 £
(https://www.ionos.co.uk/servers/vps?ar=1#packages)

This is the cheapest I could find. You know how many Helium miners are deployed. Maybe it would be good idea to target this group?

BR
Radek

[scott]

Still looks bad using that calculator(only the per 1TB storage price looks okay):

Even the storage price is high, especially before discount. Wasabi offers 1TB for $5.99 per month, with discounts for longer contracts.

In most cases using a 40% discount it’s not enough to be competitive with classic hosting and they don’t ask for a year long contract for that price (Amazon gives also good discounts if prepaid 1-3 years).

I agree, it’s problematic to compare TF discounted rates with other providers pay as you go rates, especially when the other providers also offer discounts when capacity is prepaid.

Whether or not we should even be comparing these directly from a quality/SLA perspective is a larger question. For now, I think we can at least agree that the Grid is a new technology which still needs to earn its trust. Setting a really competitive price, at least to start, is a great way to attract usage and build that trust.

Remove the 50TFT barrier on the bridge to BNB, set it to 1TFT like it’s on the inbound(there is no real world reason why it needs to be 50 except to minimise miners cashing out)

There is a simple reason for a higher fee in this direction, which is to cover the gas cost on BSC. In the other direction, the user has already covered the gas fee and all that remains is a tiny fee on Stellar. That said, I agree that 50 TFT is too high relative to typical gas costs. The devs do have a proposal to scale this based on actual current gas price.

Just to be clear, the capacity pricing is always dollar denominated, based on a (currently manually coded) TFT price. So far that’s always been $.1, meaning that there’s such a discount as you’re suggesting even now.

Looking at historic token price, even seeing it bottom out around $.01 wasn’t a sufficient discount to drive demand for consuming capacity. There’s much more that’s needed, including a nice UX, support, and sufficient testing.

We’re not as vocal about it as farming right now, but we do have ongoing efforts to onboard partners who want to utilize the Grid. Our resources are limited, so most energy is focused on growing the Grid and working out the kinks before too many users arrive. As you’re well aware, there are still very important questions to be answered, like how to handle bad actors on the user side.

Anyway, I think it’s really productive to hash out attractive dollar denominated prices for Grid capacity, so thanks for your contributions here. As someone who actually uses the Grid constantly, I can say that the current pricing feels high to me, even with the 50% discount on testnet.

[ParkerS]

Out of curiosity why are you needing a public ip for a helium miner or are you running a validation node? Miners work well under a private ip so long as you’ve forwarded 44158.

[sensorium]

Anyway, I think it’s really productive to hash out attractive dollar denominated prices for Grid capacity, so thanks for your contributions here. As someone who actually uses the Grid constantly, I can say that the current pricing feels high to me, even with the 50% discount on testnet.

So what is the action plan to get this into production?

[sensorium]

One thing I completely forgot. AWS has a bidding system! Implementing a bidding system would automatically „solve“ this issue of pricing as it would be demand-supply based.
Here it‘s maybe a bit different since it‘s contract based, but I‘m sure we can figure something out which drives usage and the system can be refined as real world data comes in from actual usage.
If I‘m not mistaken TF works on a hourly billing clock? That would mean a contract which is bidding below „normal price“ would be cancelled it there is a higher paying contract deployed on the same machine. Probably a range would be needed so that a contract can jump to a higher price if there is higher demand. Demand price probably need to be set on a machine level(to account for variations within machines), and a contract which is trying to remove an existing contract should probably have a minimim run time.
Obviously who is paying full prices would not be subject to such cancellations.

[kristof]

i’ve been reading all above, thank you so much all for your input.

Couple of remarks, pricing is independent of TFT USD price, the TFDAO sets the pricing per CU/SU/NU in USD. In the future there will be more than 1 TFGrid (we are going to support multiple regional internets), for each regional internet the local DAO decides on pricing levels because they can be different in different parts of the world.

I just did an exercise again, we have benchmarked against, digital ocean, amazon, google and OVH, for 40% discount we are competitive on all levels.

I suggest we do a GEP (Grid Enhancement Proposal) where we put the 40% discount as std for now.

About special usecases we have the channel discount, which is being implemented in TFChain right now. This means that for special usecases e.g. Presearch we can ask the DAO to approve upto 50% discount, this happens per usecase. This channel can also be send to people who create solutions on top of TFGrid.

For dedicated nodes we have another 50% discount, which I believe is great.

The pricing calculator see here.

Please do note that its hard to compare to certain specific usecases e.g.

• discounters who make certain of their products super cheap, to get you in their offering to as such then later make money in other ways.
• oversubscription, many storage offerings do oversubscription, while they say e.g. 1TB its not 1TB its a max you can use and they keep in their mind that most people won’t get there. In TFGrid case 1TB is 1 TB. 5 USD for 1 TB of quantum safe storage, which has other benefits too is probably a good price taking that into consideration.

Example machine 8 GB

see here

Our price 23 USD

Example for OVH

OVH is one of the biggest players and super aggressive in price.
We checked their cloud offering (we are still more cost effective), even on their dedicated machines we are coming out ok.

we are at 47 USD / month

Conclusion

Would like to ask support from community to go with above suggested pricing, see GEP here

Please note that the channel discount is an additional 50% discount possibility even better than what has been compared above.

[Mik]

Those numbers look great.

I have a quick question: when you say

40% discount as std for now

Does it mean users don’t need to have the corresponding TFT in their wallet to benefit the 40%? (12 months worth of TFT in account)

Thanks

[kristof]

indeed that is what I suggest for now, just to remove any barrier for people to use the grid.
maybe this is not needed and not wanted, looking forward to feedback.

[Mik]

Oh wow. OK well then clearly this gives an amazing edge for Threefold in terms of price competitiveness.

Also, here’s an idea based on your last post.

As you stated;

“discounters who make certain of their products super cheap, to get you in their offering to as such then later make money in other ways.”

Could we do this, with the 40% discount? Let me explain:

You start your TF grid journey with 40% discount without needing to “passively” stake TFT in your wallet for a 12 months worth of TFT, so as you said a 40% discount by default.

Then after a certain period, to benefit from the discount, you need to have the 12 months worth of TFT in the account to get the 40% discount.

I hope I’m being clear. Amazing what’s going on with TF these days. Thanks for your hard work @kristof