Bitcoin Forum

Some altcoin supporters have argued in the past months that "people pay now in altcoins because of the high Bitcoin transaction fee".

I've researched a little bit - no academic quality, but enough for a meaningful forum post - and concluded that that is not the case. With respect to transaction volume, Bitcoin is still dominant - with the only exceptions of Ethereum and Ripple.

I took blockchain explorers from the top 10 cryptocurrencies and counted the transactions in a reasonable time interval before 9 PM UTC. The time interval selected depended on the block interval. In the case of Bitcoin, I took one hour, with Dash, 30 minutes, with LTC, Monero and Bytecoin, 20 minutes, with ETC, NEM and Stellar, 10 minutes. and in the case of Ripple, as I didn't find a blockchain explorer I used the 24 hour volume from https://xrpcharts.ripple.com. Ripple volume is taken from this source (https://data.ripple.com/v2/stats/?start=2017-05-24T09:00:00Z&end=2017-05-24T09:59:59Z&interval=hour).

So I got the following list:


Take your own conclusions.

I perhaps will repeat this little experiment at some moment in the future.

*Ripple: The lower number are payment transactions, the higher number are all transactions (including offers created and canceled at the exchanges between Ripple's IOUs, trust set, account set (new accounts?), and "regular key" set.

Edited: Probably I have grossly understimated the Ripple volume according to Sukrim. According to that source (https://data.ripple.com/v2/stats/?start=2017-05-24T09:00:00Z&end=2017-05-24T09:59:59Z&interval=hour) there were 43569 transactions/hour in the observed interval. That would change the conclusion a bit. However I'm not updating the number still because I must figure out the difference between "payments" and "transactions" in Ripple - the "payments_count" value is close to my estimation (1210). (see my two edits here (https://bitcointalk.org/index.php?topic=1932126.msg19202004#msg19202004))

Hmm... If you look at the Tx/hr, Bitcoin seems to account for half of the overall transactions.
This is in line with the market cap dominance.
Extending this, if the artificial cap on bitcoin transactions is removed, the market cap of Bitcoin would zoom.

...

d5000

I would be interested to see an approximation (likely the best that you could do with all the price volatilities) of DOLLAR volume.  If BTC trx volume is that dominant, I would guess that $-volume of BTC trx would be even higher than your figures.

I always enjoy seeing "from the ground up analysis" of cryptos.  Keep at it!

Coinmarketcap has a decently accurate volume counting the past 24hrs in $USD I believe, and Bitcoin is listed as moving around $1B every 24 hours, and the next up move at the highest $400M over the course of 24hr, most are less than that.
I don't have a good analysis but that's a little supplement while you wait for him to get back to you.
EDIT: looking now Bitcoin is moving ~$2B while ethereum moves around $600M. So double the numbers I have listed.

Which is wrong.
You can look up metrics here https://xrpcharts.ripple.com/#/metrics or view single blocks with a tool like https://ripple.com/build/ripple-info-tool/ to verify the amount of transactions.
It would be easier to use the API call https://ripple.com/build/data-api-v2/#get-stats (https://ripple.com/build/data-api-tool/#get-stats) to get them for a period of time (https://data.ripple.com/v2/stats/?start=2017-05-24T09:00:00Z&end=2017-05-24T09:59:59Z&interval=hour is what you wanted I think? --> 44058 transactions in total during that time frame). You can also go to https://wisepass.com/ for a non-Ripple-Inc page in case you mistrust them. You can also run your own node and query that, but it is not the easiest route...

how ar epeople payign in altcoin if no altcoin are accepted anywhere? they are lying obviously, altcoin are only used to have more bitcoin this is a fact

but it's also true thta the high fee of bitcoin is currently reducing the amount of transaction that are made, i mean if the fee was far lower we would have far more TX per day guaranteed

Most altcoins are accepted and used at exchanges like Poloniex etc., so traders sending money between exchanges and to/from their wallets might make a significant part of transaction volume in any cryptocurrency (including Bitcoin).

The difference between bitcoin and altcoin or alternative coins is very big. Bitcoins have already established itself and it has even a local exchangers all over the world on which the altcoins doesnt have.Thus this give bitcoins a huge advantage over altcoins and we call that accessibility and market. And with the market pumping the price of bitcoin the holders of altcoins are selling their coins to get hold of bitcoins.

I don't deny that Bitcoin is still the king both in the number of transactions and raw volume processed

But for all your efforts, your stats don't in the least disprove the claim that "people pay now in altcoins because of the high Bitcoin transaction fee". Obviously, the point you are trying to challenge comes down to the claim that people now started to use altcoins a lot wider than before because of hefty Bitcoin transaction fees. I guess this is the new reality, and to correctly evaluate it you should use the number of transactions then and today, i.e. evaluate the change in the number of transactions, not their absolute readings as of today. For example, I for one started to use Litecoin as a value transfer vehicle since the transaction fees of this coin are really low (on the order of a few cents and below), and I also think that Litecoin is heavily undervalued, so I'm not afraid of it crashing in the interim

first of all let me tell you good job, i meant to do something like this but i was lazy, good to see someone else finally did it.

also i want to mention that all these numbers are a bit higher than their normal numbers because of the high price activity in all of these coins. the data is better gathered in different times for example from a random day each week for the past year and then be compared.

for example the number of bitcoin transactions will be lower by about 30% if you compare it with 6 months ago.

and all the altcoin transactions will be 60 to 90% lower if you compare it with the same time.

and as for etherum since it is the second rank, i believe majority of the transactions are because of the big increase in the number of ICOs that are being started on this platform and all of them need ether for investment.

Bitcoin still dominate crypto currencies world. It is like gold standard or Central Bank to others coins. You can't actually trade others coin without converts your fiat to bitcoin and use it as a reserve to others coins. The trading volume of bitcoin actually shown that it dominate the market and is not in competition with any other coin.

Yes actually your are right. Setting and building the first ever named in the market is one of the reason of dominance stock of bitcoin compare on all the coins out there. Usually people chose to used the one who set and been used by most of people so bitcoin is the no. choice of people on daily transaction. Not just that bitcoin set a development on its security and transaction system. And the development of bitcoin is not stopping because the changes it inevitable meaning to say that no one knows what improvement will happen to bitcoin in the near future. So its better to take the risk and to used bitcoin because its dominant not just for now but for the future.

Thanks! That would mean that Ripple is really a Bitcoin contender (having even more volume) and its market cap maybe isn't inflated. Ripple, however, is a different system than Bitcoin and has a completely different trust model as other cryptocurrencies. I admit that I never really investigated it in depth and I still struggle to understand what all the "ledgers" are ... ;) So for me the rational strategy is simply to believe your numbers.

We could then update the conclusion as "only Ripple and Ethereum are really used."

Do you know if the same problem applies to Stellar as well?

Edit - Important: I have still not updated the table because I looked at your source (https://data.ripple.com/v2/stats/?start=2017-05-24T09:00:00Z&end=2017-05-24T09:59:59Z&interval=hour) and there are two interesting numbers:

payment_count: 1210 (that's probably what I had counted because it's close to my estimation)
transaction_count: 43569

What is the difference here? What are "transactions" that are not payments?

Edit2: I think I figured it out: Transactions can be "payments", but also "created offers of an exchange", "canceled offers" (these two are the most common), "trust set", "account set" and "set regular key". The problem is if we wanted to compare them with Bitcoin, we would have to differentiate between "Bitcoin payment transactions" and "Bitcoin non-payment transactions" (e.g. OP_RETURN transactions like Counterparty or Factom txs). For now, what I'll do is to include both numbers in the table, but as in Bitcoin most transactions are definitively payments, then we can safely conclude that Bitcoin is used much more for payments than Ripple.


Yep, you're right, a historical comparation would be even more insightful. That's a second step I plan to do. Also to be more exact I would like to compare the daily transaction amount and not a small timeframe like I did here; for that I need charts (like the one Sukrim provided for Ripple). But for now my goal was to get a rough absolute number, to get the order of magnitude of the differences between BTC and alts.

I think a timeframe in mid-December (not too close to the holidays) would be a good point to compare to today because in these times the altcoin bubble started (with Dash and some others) and also Bitcoin blocks wern't as full as today then. Let's see. Maybe next week I'll have time for it.

@talkbitcoin: This seems also to be a good answer for you.


The stat  I was interested in is the "number of transactions", it's not the transaction volume measured in BTC currency units. This would be another interesting comparation, however. I'll look at it if I have time.

Apart from that, time intervals seem to be skewed

I don't know about other coins, but Litecoin has an average time span between blocks set to 2.5 minutes (if I'm not mistaken). So if you take 60 minutes for Bitcoin as a reference point (i.e. 6 blocks found on average), you should set time interval for Litecoin to 15 minutes (2.5×6=15). Thus the number of transactions per time interval should be lower for that coin. As to me, showing the number of transactions processed daily with reference to dates would be more informative overall than per hour stats (a chart would be even more illustrative)

Ripple is a decentralized exchange that also has a native asset (to pay fees in) called XRP.

This means transactions are often offers that are being created/updated/removed, like on any other exchange out there. This would not be possible to do on Bitcoin itself, since the blockchain there is too limited in capacity and throughput and also pecause of the PoW consensus model which allows miners to front-run.

Anyways: If you are looking at raw transaction amount, yes, systems like Ripple or Ethereum (which also has a native asset to pay for fees and lots of other ways to write transactions other than "send X to Y") it seems a bit weird to compare them to Bitcoin, which has only very limited non-payment functionalities.

With evaluating Ripple Tx volume in the "moved IOUs/XRP vs. just offered a trade" sense another difficulty is that I think xrpcharts only counts direct payments denominated in the same currency as "payment" (a simple case: Account A sends 100 XRP to account B which receives 100 XRP). There are more complex ways to also send money and users might not even see a difference directly (account A sells 90 XRP for 1 USD, sells the USD for 1 EUR, sells the EUR for 100 XRP and sends the 100 XRP to account B - all she sees is that it is cheaper that way for her to do it that way). There is also the "Exchanges" metric on https://xrpcharts.ripple.com/#/metrics which correspons to actual trades happening (so in the end also money changing owners, not just offers to buy/sell).

In the end it depends on what you want to measure:
Total capacity of a system? Total Tx it is, no matter how they are actually used.
Total amount of transactions that result in at least a very probably case of account 1 having less money and account 2 having more (not much closed to define, since you can send to yourself on a different address in most/all systems)? In Ripple you'd need to look at "Payments + Exchanges" data.
Total amount of "base case" transactions (A sends to B without touching any markets in between)? Payments it is.

I agree with your general sentiment that payments on Bitcoin are still dominating the ecosystem, even if in some cases others are handling more actual transactions currently. It would be interesting to research this kind of data for all systems in your table though, since it looks a bit weird to have so many exceptions and *** around Ripple just because they easily allow you to have more insight into differenciating transactions. They are not the only decentralized exchange system in your list... Edit: It also seems a bit weird considering the title of this thread is "transaction volume" not "payment volume". /edit

By the way: The Ripple blocks (called "ledgers") analyzed in my example are the 961 blocks from 29989105-29990065 (both inclusive)

Some more hard facts :
- Bitcoin dominance might be a reason we are still feeding greedy miners.
- Practically it will not be easier nor possible to switch over to some altcoin payments suddenly when we are into bitcoin payment based agreement.
- Probably some gamblers might have switched over to altcoin payments but I am sure they too might be finding difficulties in calculating risk/reward ratio with a new value denomination.
- Steller and bytecoins were having ~ 200% to 500% growth in recent times but not having at least 1% transactions of what bitcoin system has. How smartly devs try to fool traders.

If i understand correctly, one transaction in Bitcoin takes an hour in today`s reality. If we add to this significant system fee, it`s hard to believe that without much rebuilding bitcoin could sustain its dominant position on the future market.

Some of these points are debatable

First, it is not clear how Bitcoin dominance (which I agree with, just in case) could be the reason why we are "still feeding greedy miners". This is a vague statement in and of itself which requires further clarification and explanation. Next, I don't see any difficulty (let alone impossibility) for substituting Bitcoin with something else (another altcoin) for the very simple reason that both Bitcoin and altcoins are used primarily for trading (let's remain realists here). Strictly speaking, I don't quite understand what you mean by "bitcoin payment based agreement". Which agreement do you refer to specifically? And last but certainly not least, I myself invested some monies into Litecoin now (and am going to invest more if the price goes lower) since I consider that this coin exactly can and will likely eat away a mighty share of the Bitcoin pie in the nearest future, especially if quarrels and fights in the Bitcoin camp are set to continue (which seems to be the case)

I think the reason for such low transaction volume of altcoins is their volatility - it's a serious reason why many merchants don't want to accept Bitcoin, and volatility of altcoins is much higher, even top 10 altcoins can change their value by 20-30% in a day. If we imagine that all prices stopped moving, than people would quickly switch to alts, because Bitcoin transcations right now are slow and very expensive. But this switch might cause similar problems to alternative blockchains, because no one yet seen them under such load as Bitcoin is suffering right now.

i don't know any altcoin that has full blocks so this factor shouldn't really matter. because whether the altcoin is generating a new block every 5 seconds or every 10 minutes there will still be the same number of transactions in total.

lets say there are 100 transactions per 5 minute.
if the blocks had 1 minute intervals there would be 5 blocks each with 20 tx in them.
and if it was 5 minute interval there would be 1 block with 100 transactions in them.

although if the method you are using to retrieve the number of transactions should exclude the coinbase transaction of these blocks. because for an altcoin with lots of blocks in 1 hour there is a large, considerable amount of coinbase transactions.

I don't quite understand which factor you refer to

But whatever you might mean this certainly has nothing to do with what I meant to say. I'm just pointing out that the number of Litecoin transactions per unit of time (in this case per hour) should be higher (or lower per time interval), that's basically all. Apart from that, it doesn't matter either if the blocks are full or not, since as I got it the aim of the OP is to show whether the growth in Bitcoin prices actually made folks switch to using altcoins (as is the common assumption across the forum). And that would be interesting to find out but for that we should look at past stats too

I was mainly interested in the "base case" payment volume. But I see your point. The problem is that if I differentiate between transaction types, I cannot simply use the raw data of block explorers but would have to examine transactions in detail, because for most coins there aren't such nice charts available like for Ripple. That would take much more time. It may be worth it, though, if I find an automated way. As a long term goal I'm thinking about a web site that permits to compare different cryptocurrencies by transaction/payment volume (not trading volume/"market cap" like so many sites do).


Yes, here I agree with talkbitcoin. I selected the intervals only to have at least 6 blocks per currency and at least 10 minutes, so it was not too distorted. In my planned 2.0 version of that list I'll try to take entire days (or even 7-day averages) as measures, looking for charts or an automated way (block explorer API ...) .


Yes, that's another fact I've thought about. In v2.0, I think, I'll differentiate transactions in three groups:
- payments
- coinbase transactions
- non-payments (Bitcoin's OP_RETURN, Ripple/NEM exchange activity etc.)

Don't expect it to be ready in less than a month, though ;)

Bitcoin would have a lot more non-payment volume if more volume would be possible I guess - Mastercoin or however it calls itself these days for example or various colored coins implementations (even notification tx spam like Satoshi dice a few years back) currently are probably just priced/crowded out of limited block space. They still do exist on Bitcoin too though... so you might want to look into some more in-depth analysis of transactions in various blockchain systems.

I don't really see how coinbase transactions would be of much interest (there is just one per block and that's it?), personally I find on-chain and off-chain payment and exchange volume more interesting (ho much value gets moved in the native currency and how much trading happens off-chain in comparison?)

I only buy altcoins to later sell them for a profit in terms of bitcoin value. So this is true for me. Also, I hold off on transactions a long time and don't spend as often so my fees are only used when I actually need to pay them. What I mean by that is I send some funds to another one of my wallets but what I do is sometimes wait until every other time I get funds to send them so it's half as many inputs into the next address.

Coinbase transactions are interesting mainly to be "filtered out". Above all, in coins with small block times (under a minute) these transactions are responsible for a relatively large part of the transaction volume. That's even true more so in smaller (Bitcoin-based) coins where often most transactions are coinbase transactions - for example even in Stratis (position 11 on Coinmarketcap) 33 to 50% (https://chainz.cryptoid.info/strat/) are coinbase transactions (most blocks have 2 or 3 transactions, being one of them a coinbase tx).

How would you get the number of off-chain transactions?

Another interesting metric I just saw is probably to count the number of outputs. If I understand ring-signature/anonymous coins like Monero or Bytecoin well, then they would have a bit more volume because transactions are regularly joint together.

Ripple (like Ethereum) is account based, not UTXO based, so it is a bit hard to come up with metrics for "number of outputs". A possible metric might be "size of the current system state" (which would be all accounts, trust lines, order books etc. in Ripple and the UTXO set in Bitcoin for example).
Still not an apples to apples comparison if you compare across chains, but you could track how much this state changes over time within each system, so if transactions just change existing information or add something new.

For off-chain you could at least take a look at trading volume at various off-chain exchanges.

Good research and a real eye-opener for bitcoin haters. Bitcoin accounts almost half of the total crypto currency transactions which shows that bitcoin enjoys huge public trust. Apart from your research, the recent price surge also confirms the same. The transaction fees in bitcoin is obviously higher than other crypto currencies, but big transactors are not simply bother about the fact till the time, they enjoy the price increase and huge investment from entire earth.

Bitcoin was always the number one crypto currency and will remain the same no matter what other cryptos come and go. Boss remains the boss always.

You're right; that makes things even more complicated, but my intention with the "output based" approach is to measure how many accounts/addresses are receiving "money" from a set of transactions. That should also be analyzable with account-based currencies.

If I analyze the system state size, then there should be a clear differentiation between payment and non-payment items.


Can you give me an example for an off-chain exchange? I'm not aware of that (or are you talking about, for example, Bitcoin IOUs on Ripple or Nxt?)


No, that cannot concluded from my "research" (until now, it's only a forum post with perhaps over-average elaboration ;) ) because I only analyzed 10 currencies. As there are over 3000 altcoins, I think Bitcoin has far less than 50% of the transaction number of the cryptocurrency ecosystem  - only coinbase transactions should make another 10.000/hour, as most altcoins have far smaller "block time targets".

Kraken, Bitstamp, Bitfinex... they all are only exchanging Bitcoin IOUs, not actual BTC. I'm not 100% sure if something like RCL or Bitsquare should be pointed out, maybe by introducing a "publicly auditable off-chain exchange" column? After all I would also be more interested in coins where I can actually at least see and verify all offers and trades and not having to hope that Poloniex or Shapeshift don't defraud me...


Nope, looking at the raw numbers for example, Ripple has about 2 times more transactions than all other analyzed coins (including Bitcoin) combined, Bitcoin only has ~1/4 of all transactions of the analyzed systems. This doesn't mean that one is better than the other, it just means that apparently Ripple manages to actually profit from its cheaper/faster consensus algorithm by offering additional capacity that actually gets utilized. Other coins seem to struggle a bit more with adoption, since they have sometimes quite high theoretical TPS limits, but in practice their actual transaction volume gets even dwarfed by Bitcoin with its (currently) 1 MB/10min blocks.

...

d5000

Great work.

Another issue that might be worthy of research would what MERCHANTS accept any of the "Alts" for payment.  I know of ONE that takes Monero and will likely take Ethereum soon.  But, that's it!  (Note that I am no expert.)

We all know that there are over 100 merchants that take BTC (and I KNOW providentmetals.com takes BTC as I have done business with them), but who would take Alts?  And in what sort of volume.

For that matter, who (other than exchanges, maybe) would give CA$H for Alts?

Truly we are entering a brave new world of payments!  Research like many of you above are doing is fantastic work, adding value to our lives.

OK, so you're referring to "normal"/"centralized exchange" volume. Well, that number is already tracked by multiple sites (coinmarketcap et al.). That are not numbers I'm particularly interested in, as every sh*tcoin can be pumped on Poloniex and can get high volume numbers. But as these numbers are easily available I can add them to contrast the number with the transaction volume.

What would be more interesting for me would be off-chain wallet usage but I don't think that data is publicly available.


Yes, that is more in scope with my analysis, but it is pretty difficult to investigate. An idea could be to combine the acceptance by payment processors like Coinpayments (https://coinpayments.net) with the search volume of word combinations like "XRP accepted", "BTC accepted" etc. but it is much more difficult to track.

For now, my idea for a second version of this "study" would have the following characteristics:
- some coins more (top 20 or even top 50)
- daily volume in transaction number (if someone knows easy-to-analyze charts for the top coins, feel free to mention them in a comment)
- a historical comparation (December 2016 to a date near today)
- if possible: transaction types (payment vs. non-payment, so I won't discriminate poor Ripple again ;D ). My idea for now is to differentiate between transactions that involve only one key-pair or account (non-payment, e.g. OP_RETURN and Ripple offers) to those that have more than 2 key-pairs/accounts involved (payments and on-chain exchange between two parties, like in the Ripple case). That should be doable.
- contrast the transaction volume with exchange volume tracked by Coinmarketcap.

More not because otherwise it will be vapourware almost surely ;)

It would also be great if you'd manage to write this into a script so it is possible to let it run at a later date/on demand/continuously on a website (ther is a real need for websites listing hard facts other than current exchange rate, similar to what coingecko does).

Bonus points for running it against actual blockchain-native client APIs (bitcoind, geth etc.) instead of parsing web APIs.

A script-based approach is in my plans, so your first point probably will make it into version 2.

Unfortunately, at this moment for me it is difficult to impossible to run full nodes of 20 or more cryptocurrencies. So point 2 will have to wait unless I find a way to convert that into a business like coingecko (it's certainly an inspiration, but I don't know why they haven't integrated that point - transaction/payment volume - yet).

OK, I'm now preparing the second version of this study.

I will write a little script that checks block explorers and returns the 24h transaction count for a coin. For that to work, I need blockchains explorers with a working API that provide me the following methods, given a block height:
- timestamp of the block
- transaction list OR transaction count of the block

or alternatively:
- transaction count of a freely definable or 24 hour time interval

For the following coins of the current Top20 on Coinmarketcap, I have already found blockchain explorers that met my criteria:

Bitcoin, Ethereum, Ripple (XRP), Litecoin, Ethereum Classic, Dash, Monero, ZCash, Bytecoin, Dogecoin

The following ones seem to have a working block explorer with API but I couldn't get them to work or they hung:

AntShares, BitShares, Steem

For the following coins I didn't find block explorers which meet my criteria (I need the functions or I can't include them):

NEM, IOTA, Stratis, BitConnect, Waves, Siacoin.


If you know a working solution for the coins in orange or red, I'll be glad to take it into account - simply post it in the comments.

If I don't find block explorers with the required API methods for these coins, I will include them with a rough estimation, indicating that their community is so small that they're not able to provide a decent block explorer.

Of course the bitcoin will be the dominance of everyone's choice. Because bitcoin has a very profitable investment assets. Many people are turning to bitcoin for investment. All coin just as a complement to add bitcoin value. So too with gold. I think even though many people are investing in gold. But it can not be denied that bitcoin has a much higher value compared to gold. So I think for now bitcoin is still the most dominance in the select. And I hope it will continue for the better in the long term

This is valid, exchange charges for bitcoin are low from each other digital money. This has been the situation from the introduction of bitcoin and before the birthplace of other digital currencies. This is one of the purpose behind the accomplishment of bitcoin to such an extent. Other digital forms of money are not that much valued and have low overall revenues than bitcoin.

It seems as time goes on, Bitcoin loses more and more transaction volume share to ICOs/altcoins. It is to be expected as programmers around the world are learning how to do this stuff and want to make and promote their own brand. The problem arises when they get greedy, which almost always happens, and as a result everyone loses. We need some sort of closure to this madness, and hopefully soon. :P

It's just that Bitcoin has been around for many years now since the fees are reasonable that's why it's transaction and volume overwhelms altcoin But today, I think some people are starting to move to altcoin because of the increasing transaction fees in Bitcoin.

Yet his project looks like it would be excellent work, even if only, say, just five cryptos are chosen at first.  Could always work others in at a later point, I presume there would be no real rush.

I'd like to see the below done (this is just my *best guess* as to most USED cryptos in actual commerce):

Bitcoin
Litecoin
Monero
Ethereum
ZCash  (although I read something very negative about the controllers of ZCash today at Zero Hedge)

Later I would like to maybe see:

ZCoin  (apparently without that above control issue, but, hey, I don't know)
IOTA   (apparently it's an interesting idea, but weak technically ?))

Just my thoughts, FWIW...

Bitcoins are extremely slow when it comes to transactions and at the same time they demand such high transaction fee that its simply stunning that such a huge amount of population is not switching to another search engine.

well I do think that we need a fork to further increase the digits otherwise notvafter a long time.. it will be hard for bitcoins to dominate..

But then again bitcoins is the most famous one and the most powerful one but it should improve to be able to sustain this.