why does this forum keep removing my double spacing between sentences?

It’s not the forum: it’s your browser. HTML layout requires multiple consecutive space characters to be collapsed into one.
Also, The Mac is not a typewriter…

You mean to say the moron who designed html didn't allow for two spaces between words? "& n b s p ;" is 6 bytes, a space is one. What a waste. And I have no idea what you're referring to with that book, I'm not going to read the whole book. And I don't use a Mac.

Based on my past typographical experience (physical typesetting and computer font design), double-spacing only makes sense with monospaced fonts, but is entirely redundant with proportional fonts of any kind.
Sentence-spacing is catered for differently to word-spacing, so your extra space just makes things look weird.

I do sympathise because I learned to type when my mum was learning to type on an original typewriter in the 70s (Pitmans course) and was originally a double-spacer too. But you've been wasting your time for at least a couple of decades.

Like with many things nowadays, your solution is to stop wanting redundant things and just press fewer keys. Everyone wins.

---

I haven't seen more than 360 credits per task

Sounds low. Are your machines throttled (other than when they’re running the benchmarks)? If not, the Ryzens and fast i7s should be doing better than that. (For comparison, 360 is about the peak for my 4th-⁠gen i5s, which have lower measured performance than many of your CPUs.)

I'm currently using an I3-8350 and getting 550-600 per 8hr task per core. Definitely low

---

I haven't seen more than 360 credits per task

Sounds low. Are your machines throttled (other than when they’re running the benchmarks)? If not, the Ryzens and fast i7s should be doing better than that. (For comparison, 360 is about the peak for my 4th-⁠gen i5s, which have lower measured performance than many of your CPUs.)

I'm currently using an I3-8350 and getting 550-600 per 8hr task per core. Definitely low

I'm doing tasks in roughly 2 hours and getting @65 to 70 credits per task

I'm doing tasks in roughly 2 hours and getting @65 to 70 credits per task

It seems that credits don’t scale quite linearly with time, at least at the low end; I’ve seen this too when I’ve reduced to a shorter run time. (You would expect more than four times as much credit from your i7 if your target run time was four times as long.)

I'm doing tasks in roughly 2 hours and getting @65 to 70 credits per task

It seems that credits don’t scale quite linearly with time, at least at the low end; I’ve seen this too when I’ve reduced to a shorter run time. (You would expect more than four times as much credit from your i7 if your target run time was four times as long.)

Exactly!! Otherwise why run them that long? The Project needs to rethink the credits awarded or stop giving us the option for shorter units that get proportionatly more credits. If I'm running the task for 2 hours and find ie 10 results and you run the same task for 8 hours and find 15 results then I'm missing results and that could be REALLY REALLY bad if everyone only ran the tasks for 2 hours.

Based on my past typographical experience (physical typesetting and computer font design),

So you're one of those fools who makes fonts where LlIi1 are similar or even identical? You can assume which it is in a word/sentence, but not if it's a code/password to type in.

double-spacing only makes sense with monospaced fonts, but is entirely redundant with proportional fonts of any kind.

The font being proportional makes no difference here. I want the space between sentences to be double the space between words, for clarity of reading.

Another thing this forum does is mess with my line breaks. It inserts two automatically after your quoted text, so I have to remember not to put any in. Primegrid inserts one. Some forums insert none. ARGH!

Sentence-spacing is catered for differently to word-spacing, so your extra space just makes things look weird.

No it isn't, it's not done automatically. I'm seeing the same gap between words and sentences in this forum.

I do sympathise because I learned to type when my mum was learning to type on an original typewriter in the 70s (Pitmans course)

Boots, late 70s.

Like with many things nowadays, your solution is to stop wanting redundant things and just press fewer keys. Everyone wins.

"Fewer" is a redundant word, just use "less".

---

I'm doing tasks in roughly 2 hours and getting @65 to 70 credits per task

Hey Mikey, have you got a CPU as fast as mine? Just installed a Ryzen 9 3900XT! Stupid Currys were selling the XT cheaper than the X, even though it's identical with 10% more clock.

---

I'm doing tasks in roughly 2 hours and getting @65 to 70 credits per task

It seems that credits don’t scale quite linearly with time, at least at the low end; I’ve seen this too when I’ve reduced to a shorter run time. (You would expect more than four times as much credit from your i7 if your target run time was four times as long.)

Maybe they don't want you running it longer? Maybe they want to feel sorry for the slow CPUs?

---

Exactly!! Otherwise why run them that long? The Project needs to rethink the credits awarded or stop giving us the option for shorter units that get proportionatly more credits. If I'm running the task for 2 hours and find ie 10 results and you run the same task for 8 hours and find 15 results then I'm missing results and that could be REALLY REALLY bad if everyone only ran the tasks for 2 hours.

I don't think they give you an option as such. You have to fiddle with config files to do so, and most people don't. They've set the default to 8 hours as that produces about how many results they need. A scientist once said they don't need the same number of results from every task, as long as the average is what they need.

---

why does this forum keep removing my double spacing between sentences?

It’s not the forum: it’s your browser. HTML layout requires multiple consecutive space characters to be collapsed into one.

Actually, in the Nexus mod forums the double space works. So it's not the fault of html or my browser, but this forum. Again.

---

I'm doing tasks in roughly 2 hours and getting @65 to 70 credits per task

Hey Mikey, have you got a CPU as fast as mine? Just installed a Ryzen 9 3900XT! Stupid Currys were selling the XT cheaper than the X, even though it's identical with 10% more clock.

NO I do not, my 2 desktop Ryziens are 1920X's which I use 23 cores of while my laptop which has a Ryzen 7 4800H of which I only use 10 cores of so i can use it for other things like typing this etc.

Exactly!! Otherwise why run them that long? The Project needs to rethink the credits awarded or stop giving us the option for shorter units that get proportionatly more credits. If I'm running the task for 2 hours and find ie 10 results and you run the same task for 8 hours and find 15 results then I'm missing results and that could be REALLY REALLY bad if everyone only ran the tasks for 2 hours.

I don't think they give you an option as such. You have to fiddle with config files to do so, and most people don't. They've set the default to 8 hours as that produces about how many results they need. A scientist once said they don't need the same number of results from every task, as long as the average is what they need.

I chose the 2 hour setting to run my tasks when I setup my account on Rosetta and I've just never changed it.

If a two hour task completes 10 models, then an eight hour task is more likely to complete about 40 models, not 15.

Note that not all tasks can complete in two hours. With such a short runtime preference, you are more likely to see tasks running longer than the preference. When you look at credits, you really must consider the amount of actual CPU time, not the number of work units, and not just the runtime preference.

There are no "missing results". So, set your preferences in a way that works for you and your machine.

If you use Dr. Baker's analogy of exploring a planet's surface for the highest or lowest elevation on the planet, then each model is one of the explorers. They start their exploration from a random point on the planet. When a work unit has enough time to begin another model, that next model will be started at another random point on the planet, with no regard to the first model or what it found. If you drop 10,000 explorers on the planet, your success in finding the true highest or lowest elevation would essentially be proportional to the surface area of the planet. If 10,000 explorers is adequate for Mars, you might need 100,000 for Saturn. So, when they feel they have a Saturn-sized protein for study, they might create more work units. But, as you point out, they have no way to predict exactly how many models will result. If they approach the end of work units coming back in and still only have 80,000 results, then they create more work units to obtain the 100,000 results desired.

Having said that, once they see the results, they can sometimes give hints to future explorers, or essentially drop more of them near the Himalayas. So they might create a secondary batch of work units, which are designed to concentrate the focus based on what was learned on the first round.

---

Rosetta Moderator: Mod.Sense

Based on my past typographical experience (physical typesetting and computer font design),

So you're one of those fools who makes fonts where LlIi1 are similar or even identical? You can assume which it is in a word/sentence, but not if it's a code/password to type in.

You make a good point about the risks in font design, but seeing as your example produces different and highly distinct characters on my screen, it's an exercise in distinguishing good font design from bad. If it isn't the same for you, you're, by definition, using a badly designed font. So, no, the period in the early/mid 90s, when I moved on to computer fonts and beta-testing programmes that could design them, did not make me one of the fools you're talking about.

double-spacing only makes sense with monospaced fonts, but is entirely redundant with proportional fonts of any kind.

The font being proportional makes no difference here. I want the space between sentences to be double the space between words, for clarity of reading.

This is a good example of wanting things that don't actually look good. In the original hot metal days this would be the difference between half the font size (en-space, letter n) and the full font size (em-space, letter m) . With proportional fonts individual letters have their own tracking and kerning info attached, word-spacing is typically en-sized, with sentence ends typically (and I'd suggest visually sensibly) en-quad sized, which is 1.5x en-spacing (or 0.75 em-spacing) because the space above the full-stop already includes sufficient white-space. Anything more and you get rivers of white-space running through the text, which is visually ugly/repulsive. More importantly, I think, it's one key-press rather than two and the context will call from the font the appropriate inbuilt font-designed spacing.

You mention code; in that instance, it makes no sense to use a sans-serif proportional font that invites confusion. A monospaced serif'd font like Courier, that doesn't use character codes, solves the issue at source (sic)

tl;dr let the font handle it because spacing and it's application is designed into the style and can/does vary between type-styles

Sentence-spacing is catered for differently to word-spacing, so your extra space just makes things look weird.

No it isn't, it's not done automatically. I'm seeing the same gap between words and sentences in this forum.

I'm not sure what font you're using on your screen, but if it isn't different you either need to change your font to something better or get your eyes tested.

I know you'll like this next bit. I looked up something while writing this message to double-check my aging and failing memory and came up with this quote:

Much has changed along the journey from typewriters to setting type on computers. Still, there are a number of typewriting conventions that are no longer relevant but which stubbornly refuse to go away. At the top of this list is the practice of putting two spaces between sentences. Forget about tolerating differences of opinion: typographically speaking, typing two spaces before the start of a new sentence is absolutely, unequivocally wrong.

---

If a two hour task completes 10 models, then an eight hour task is more likely to complete about 40 models, not 15.

Note that not all tasks can complete in two hours. With such a short runtime preference, you are more likely to see tasks running longer than the preference. When you look at credits, you really must consider the amount of actual CPU time, not the number of work units, and not just the runtime preference.

There are no "missing results". So, set your preferences in a way that works for you and your machine.

If you use Dr. Baker's analogy of exploring a planet's surface for the highest or lowest elevation on the planet, then each model is one of the explorers. They start their exploration from a random point on the planet. When a work unit has enough time to begin another model, that next model will be started at another random point on the planet, with no regard to the first model or what it found. If you drop 10,000 explorers on the planet, your success in finding the true highest or lowest elevation would essentially be proportional to the surface area of the planet. If 10,000 explorers is adequate for Mars, you might need 100,000 for Saturn. So, when they feel they have a Saturn-sized protein for study, they might create more work units. But, as you point out, they have no way to predict exactly how many models will result. If they approach the end of work units coming back in and still only have 80,000 results, then they create more work units to obtain the 100,000 results desired.

Having said that, once they see the results, they can sometimes give hints to future explorers, or essentially drop more of them near the Himalayas. So they might create a secondary batch of work units, which are designed to concentrate the focus based on what was learned on the first round.

This is a terrific explanation, thank you. Based on this explanation I think I'm going to bump my WU processing time (on my dedicated 24/7 zero cache machines) from 24 hours to 36.

I'm doing tasks in roughly 2 hours and getting @65 to 70 credits per task

Hey Mikey, have you got a CPU as fast as mine? Just installed a Ryzen 9 3900XT! Stupid Currys were selling the XT cheaper than the X, even though it's identical with 10% more clock.

NO I do not, my 2 desktop Ryziens are 1920X's which I use 23 cores of while my laptop which has a Ryzen 7 4800H of which I only use 10 cores of so i can use it for other things like typing this etc.

I'm terribly sorry but I'm gloating. Christ it's warm in here [places hand over one of the exhaust fans].

---

If a two hour task completes 10 models, then an eight hour task is more likely to complete about 40 models, not 15.

Note that not all tasks can complete in two hours. With such a short runtime preference, you are more likely to see tasks running longer than the preference. When you look at credits, you really must consider the amount of actual CPU time, not the number of work units, and not just the runtime preference.

There are no "missing results". So, set your preferences in a way that works for you and your machine.

If you use Dr. Baker's analogy of exploring a planet's surface for the highest or lowest elevation on the planet, then each model is one of the explorers. They start their exploration from a random point on the planet. When a work unit has enough time to begin another model, that next model will be started at another random point on the planet, with no regard to the first model or what it found. If you drop 10,000 explorers on the planet, your success in finding the true highest or lowest elevation would essentially be proportional to the surface area of the planet. If 10,000 explorers is adequate for Mars, you might need 100,000 for Saturn. So, when they feel they have a Saturn-sized protein for study, they might create more work units. But, as you point out, they have no way to predict exactly how many models will result. If they approach the end of work units coming back in and still only have 80,000 results, then they create more work units to obtain the 100,000 results desired.

Having said that, once they see the results, they can sometimes give hints to future explorers, or essentially drop more of them near the Himalayas. So they might create a secondary batch of work units, which are designed to concentrate the focus based on what was learned on the first round.

Thankyou for a well written explanation. It's nice to know how things work, so we can set settings in a way to maximise productivity.

---

You make a good point about the risks in font design, but seeing as your example produces different and highly distinct characters on my screen, it's an exercise in distinguishing good font design from bad. If it isn't the same for you, you're, by definition, using a badly designed font.

I use the standard Windows fonts. Ariel or Times New Roman. And evn if I choose good fonts, tat doesn't heklp when I see printed papers from other people. There is no excuse for any font to ever have any character similar. When I learned to read and write at school, the capital eye had a lid and two feet. The number one had a peaked hat to the left. The lower case ell had a foot to the right. No confusion at all. The most common confusion I see is capital eye and lower case ell both just being a straight line.

This is a good example of wanting things that don't actually look good. In the original hot metal days this would be the difference between half the font size (en-space, letter n) and the full font size (em-space, letter m) . With proportional fonts individual letters have their own tracking and kerning info attached, word-spacing is typically en-sized, with sentence ends typically (and I'd suggest visually sensibly) en-quad sized, which is 1.5x en-spacing (or 0.75 em-spacing) because the space above the full-stop already includes sufficient white-space. Anything more and you get rivers of white-space running through the text, which is visually ugly/repulsive. More importantly, I think, it's one key-press rather than two and the context will call from the font the appropriate inbuilt font-designed spacing.

I disagree, 1.5x en is not enough to seperate sentences visually. I add another space without any thought and it looks far easier to read.

I'm not sure what font you're using on your screen, but if it isn't different you either need to change your font to something better or get your eyes tested.

This is how I see your text.I think that's Ariel. Notice the gap between the fullstop and the capital eye is the same as the space between each of your words. And I'm not including the gap above the fullstop, as that's not a gap, it's over it, not to the right of it.

I know you'll like this next bit. I looked up something while writing this message to double-check my aging and failing memory and came up with this quote:

Much has changed along the journey from typewriters to setting type on computers. Still, there are a number of typewriting conventions that are no longer relevant but which stubbornly refuse to go away. At the top of this list is the practice of putting two spaces between sentences. Forget about tolerating differences of opinion: typographically speaking, typing two spaces before the start of a new sentence is absolutely, unequivocally wrong.

Whoever wrote that is stupid enough to think a fact and an opinion are the same thing. People have different driving styles, different writing styles, different gardening styles, etc. I like my text to have a bigger gap between sentences, it's easier to read as you clearly see where a sentence stops and starts. Computers should allow for different styles to be used.

Anyway, why should html need this ampersand nbsp nonsense? Why isn't the space character a character like any other? This is as bad as Windows getting confused when they invented long filenames and allowed spaces. Nobody ever thinks about things carefully or looks ahead to future changes.

---

This is a terrible exchange in the number-crunching forum - apologies to everyone else

You make a good point about the risks in font design, but seeing as your example produces different and highly distinct characters on my screen, it's an exercise in distinguishing good font design from bad. If it isn't the same for you, you're, by definition, using a badly designed font.

I use the standard Windows fonts. Ariel or Times New Roman. And even if I choose good fonts, that doesn't help when I see printed papers from other people. There is no excuse for any font to ever have any character similar. When I learned to read and write at school, the capital eye had a lid and two feet. The number one had a peaked hat to the left. The lower case ell had a foot to the right. No confusion at all. The most common confusion I see is capital eye and lower case ell both just being a straight line.

Some characters are very similar with sans serif fonts. Some sans serif fonts are better for clarity than others.
I've just had a mess around in MS Word and Calibri is slightly more distinct than Arial, for example, but seems a little confused to me as it uses serifs on number 1 but not on any letters or any other numbers.
If people send you documents that contain code where you want to be clear about what the letters are, they shouldn't use a sans serif font to print them out.

What you're talking about when you were at school is that you learned about serif'd fonts - the twiddly bits on numbers & letters are known as serifs. If that's what you want, you need to change your screen font to a serif'd one like Times New Roman, though maybe Century or Georgia would suit you better. But, you know, computers can handle multiple fonts, so it's not as one-dimensional as a person's writing. That's kind of like the point.

But to warn you, the reason screen fonts are sans serif is because serif'd fonts become really hard to read over time, with all those twiddly bits you need to see upsetting several points on every individual letter of every word of every sentence of every paragraph on every page. I <really> wouldn't recommend it.

I disagree, 1.5x en is not enough to separate sentences visually. I add another space without any thought and it looks far easier to read.

You can disagree if you like but, going by the quote I posted which I very much agree with, you're wrong and just because you have a different opinion doesn't mean anyone should pay a second's worth of attention to it - lol.
Fortunately, computers can handle multiple fonts and if you want to use a serif'd or monospaced one, you can change it to your preference. Do that.

This is how I see your text.I think that's Arial. Notice the gap between the fullstop and the capital eye is the same as the space between each of your words. And I'm not including the gap above the fullstop, as that's not a gap, it's over it, not to the right of it.

Thanks. I hate to tell you this, but the spacing between words and after the fullstop in your screenshot is different. Specsavers it is, then.
And writing that you aren't counting the space above the fullstop doesn't take away from the fact it's there and that area adds to the distinction from the end of one sentence and the start of another.
What's actually happening here is you're desperately hanging onto a convention that was learned several decades ago on a technology that no longer exists (to all intents and purposes) to a character style that no longer exists (except when using monospaced fonts) and demanding conformance to that non-existing (and, let's be honest, nonsensical in the current context) standard.
I learned the same way as you. I know exactly where you're coming from. But in answer to your request to cater for a meaningless redundancy, the answer can only be "No, not ever". And also "shut up" and/or "tell someone who's interested, because it's not me".

I know you'll like this next bit. I looked up something while writing this message to double-check my aging and failing memory and came up with this quote:

Much has changed along the journey from typewriters to setting type on computers. Still, there are a number of typewriting conventions that are no longer relevant but which stubbornly refuse to go away. At the top of this list is the practice of putting two spaces between sentences. Forget about tolerating differences of opinion: typographically speaking, typing two spaces before the start of a new sentence is absolutely, unequivocally wrong.

Whoever wrote that is stupid enough to think a fact and an opinion are the same thing. People have different driving styles, different writing styles, different gardening styles, etc. I like my text to have a bigger gap between sentences, it's easier to read as you clearly see where a sentence stops and starts. Computers should allow for different styles to be used.

From Wikipedia:
Ilene Strizver, founder of the Type Studio, says "Forget about tolerating differences of opinion: typographically speaking, typing two spaces before the start of a new sentence is absolutely, unequivocally wrong"
The Complete Manual on Typography (2003) states that "The typewriter tradition of separating sentences with two-word spaces after a period has no place in typesetting" and the single space is "standard typographic practice"
The Elements of Typographic Style (2004) advocates a single space between sentences, noting that "your typing, as well as your typesetting, will benefit from unlearning this quaint [double spacing] Victorian habit"
Lol

The difference between facts and opinions isn't really relevant here. They're all opinions/preferences. Some carry more weight than others and readers can evaluate those views as they see fit.
Yours isn't so much an opinion, but a learned convention that certainly applied at the time we picked it up and for many decades since, but now has no relevance whatsoever beyond doggedly hanging on to that convention after its purpose disappeared for... no reason at all bar habit. Hardly anyone below the age of maybe 30 even knows the convention existed. And eventually, both of us will die, taking our useless information with us. Harsh but true.

And, of course, computers do allow for multiple typefaces. Pick one

Anyway, why should html need this ampersand nbsp nonsense? Why isn't the space character a character like any other? This is as bad as Windows getting confused when they invented long filenames and allowed spaces. Nobody ever thinks about things carefully or looks ahead to future changes.

Why does hypertext markup language use codes for formatting? Really?
I don't know. I looked up this thread on my phone after looking at it on my desktop and I could see the question just the same, but I still wasn't sure of the answer.
Maybe someone else can help... or maybe you already worked it out because you knew the answer before performatively asking it. You're not that dumb. Nor am I.

---

If a two hour task completes 10 models, then an eight hour task is more likely to complete about 40 models, not 15.

Note that not all tasks can complete in two hours. With such a short runtime preference, you are more likely to see tasks running longer than the preference. When you look at credits, you really must consider the amount of actual CPU time, not the number of work units, and not just the runtime preference.

There are no "missing results". So, set your preferences in a way that works for you and your machine.

If you use Dr. Baker's analogy of exploring a planet's surface for the highest or lowest elevation on the planet, then each model is one of the explorers. They start their exploration from a random point on the planet. When a work unit has enough time to begin another model, that next model will be started at another random point on the planet, with no regard to the first model or what it found. If you drop 10,000 explorers on the planet, your success in finding the true highest or lowest elevation would essentially be proportional to the surface area of the planet. If 10,000 explorers is adequate for Mars, you might need 100,000 for Saturn. So, when they feel they have a Saturn-sized protein for study, they might create more work units. But, as you point out, they have no way to predict exactly how many models will result. If they approach the end of work units coming back in and still only have 80,000 results, then they create more work units to obtain the 100,000 results desired.

Having said that, once they see the results, they can sometimes give hints to future explorers, or essentially drop more of them near the Himalayas. So they might create a secondary batch of work units, which are designed to concentrate the focus based on what was learned on the first round.

Thank you very much for the explanation, it helps alot!!!