HS3 Paddington to Newport

[Mgameing123]

Stopping at both Old Oak Common and Heathrow would be a bit much, but in any case it doesn't change the fundamental logic. Heathrow to Reading is still a run of 35km, more than enough to utilise top speeds well in excess of 140mph. If you stop at Old Oak instead its ~50km.

It is alot but we got to remember these are some key transport hubs that will generate alot of riders.

 

[Meerkat]

How fast does HS2 go in tunnels equivalent to those needed between Paddington and Reading?
Just looked at a map and it’s difficult to see much of it all that wouldn’t at least have to go in cut and cover! It’s never going to make a business case….

 

[HSTEd]

How fast does HS2 go in tunnels equivalent to those needed between Paddington and Reading?
Just looked at a map and it’s difficult to see much of it all that wouldn’t at least have to go in cut and cover! It’s never going to make a business case….

There is no hard limit on speeds that can be achieved in tunnels, and indeed HS2 has smaller tunnel cross sections (imposing various operational and speed limitations) than found in schemes abroad. Overall the higher the speed, the larger the tunnel cross section. It is not obvious, however, to what extent tunnel diameter is a major driver of cost.

But if you want a 360kph train in a tunnel, you can certainly have one.

 

[Meerkat]

There is no hard limit on speeds that can be achieved in tunnels, and indeed HS2 has smaller tunnel cross sections (imposing various operational and speed limitations) than found in schemes abroad. Overall the higher the speed, the larger the tunnel cross section. It is not obvious, however, to what extent tunnel diameter is a major driver of cost.

But if you want a 360kph train in a tunnel, you can certainly have one.

I was asking because a GW tunnel would be very similar to the OOC - Ruislip tunnel

 

[Mgameing123]

I was asking because a GW tunnel would be very similar to the OOC - Ruislip tunnel

This tunnel would go from Old Oak Common to a bit outside Heathrow on the M4 where the line just follows the M4 corridor to reduce noise pollution or it would go south of Windsor.

 

[AlastairFraser]

This tunnel would go from Old Oak Common to a bit outside Heathrow on the M4 where the line just follows the M4 corridor to reduce noise pollution or it would go south of Windsor.

M4 corridor is best. South of Windsor would be very difficult for several reasons - a lot of the land around Datchet/Old Windsor is floodplain which regularly spends a lot of time underwater, and you would have to thread through a narrow corridor between Legoland and Windsor Great Park (which is part of the Royal Estate, attached to Windsor Castle!).

South of Windsor Great Park is Ascot, so the M4 or M3 corridors are the only real viable corridors west/south west out of London.

 

[Meerkat]

The M4 wanders all over the place and isn’t exactly through open countryside! By the time you sorted all the issues you may as well just build a straight tunnel.

 

[AlastairFraser]

The M4 wanders all over the place and isn’t exactly through open countryside! By the time you sorted all the issues you may as well just build a straight tunnel.

Most of the M4 alignment would be fine for a viaduct. A tunnel of that length would make the scheme completely financially untenable (and not commensurate with the level of traffic - a Phase 1 tunnel could be arguable for HS2 at a stretch, because it could carry London to West Mids, North West, Yorkshire and Scotland traffic).
South Wales, the South West, the Thames Valley and the West Country aren't really comparable traffic sources.

 

[Northumbriana]

I'd just build a new line from Old Oak to Reading via a station just north of Heathrow connecting with either those automated pod thingys or a better internal transit system. A new Severn tunnel and a new line from Cardiff to Swansea.

After that, rebuild existing tracks to take double decker trains to increase capacity or just build new tracks alongside existing ones.

 

[AlastairFraser]

I'd just build a new line from Old Oak to Reading via a station just north of Heathrow connecting with either those automated pod thingys or a better internal transit system. A new Severn tunnel and a new line from Cardiff to Swansea.

After that, rebuild existing tracks to take double decker trains to increase capacity or just build new tracks alongside existing ones.

Build new tracks west from Reading up the Thames gorge section towards Didcot and Swindon? How well do you know the area?

Regardless of the engineering difficulty of widening the Reading to Swindon section of the Great Western Main Line/GWML (at 10-11am on a Monday which I'd classify as off peak, there are 19tph using the GWML from Reading west in both directions, so would need widening for any uplift), to get any significant benefits for capacity and journey times, you'd need to build a line as far as Wootton Bassett Jcn (otherwise you'll only be marginally faster and get marginally more capacity than the existing GWML.)

Edit - the one area I do agree with on-line widening/upgrades is Cardiff to Swansea, a new line wouldn't really efficiently serve the major settlements here, and Cardiff to Swansea traffic alone isn't really enough to justify HSR for this section.

 

[HSTEd]

I think the only way we can fill this line is to go to the major load centres and serve them in the style of a Japanese Shinkansen.

A bypass line like HS2 will just end up almost entirely empty and will be little more than a giant millstone around the neck of the railway.

 

[AlastairFraser]

I think the only way we can fill this line is to go to the major load centres and serve them in the style of a Japanese Shinkansen.

A bypass line like HS2 will just end up almost entirely empty and a giant millstone around the neck of the railway.

Well, my plan posted before in this thread could be summarised as building a HSL following the M4 on a viaduct most of the way, but with LGV-style spurs to the existing GWML, to enable some trains to serve existing larger stations (apart from at Heathrow, which would be directly served).

So you would be able to significantly speed up longer services towards Cardiff, and increase capacity for existing load centres.

 

[HSTEd]

Well, my plan posted before in this thread could be summarised as building a HSL following the M4 on a viaduct most of the way, but with LGV-style spurs to the existing GWML, to enable some trains to serve existing larger stations (apart from at Heathrow, which would be directly served).

So you would be able to significantly speed up longer services towards Cardiff, and increase capacity for existing load centres.

The spur based approach can have the major disadvantage that it becomes impractical for each train to serve more than one of the load centres like that.

That means we won't be able to have attractive service densities to each destination without running a lot of short trains that are expensive to run (per seat).
It also doesn't do anything much for existing capacity limitations since you will just create bottlenecks at the major stations.

 

[AlastairFraser]

The spur based approach can have the major disadvantage that it becomes impractical for each train to serve more than one of the load centres like that.

That means we won't be able to have attractive service densities to each destination without running a lot of short trains that are expensive to run (per seat).
It also doesn't do anything for existing capacity limitations since you will just create bottlenecks at the major stations.

1.) In general, yes, but the major load centres could be served by dual easterly/westerly spurs.

For example, my Reading spur plan was an easterly spur following the A329(M)/A3290 into central Reading, and a westerly spur off the Berks and Hants line about half a mile east of Theale station (this is affordable, because the Berks and Hants line intersects the M4 at a location with plenty of land for a cheap and short western spur).

So trains can leave and rejoin the line at several locations to serve multiple major load centres, without too much of a time penalty)

2.) It depends how many of those new services you run through major load centres east of Bristol. Both the former county of Avon and South Wales could probably justify at least 1tph 12 car high speed service, non-stop from Heathrow as far as Newport for South Wales, or Bristol Temple Meads for former Avon.
So you could run those all the way up the HSR without touching the existing lines, as far as Barton Hill in Bristol with a Bristol Centre spur, or Magor for the South Wales services. This demand would grow over time, as people would no doubt switch from the M4/coaches with the journey time differential and new capacity depressing rail ticket prices.

Beyond that, you can do some localised widening and additional platforms at the bottleneck stations you described, without necessarily widening the railway on the most difficult sections.

 

[SynthD]

I think the only way we can fill this line is to go to the major load centres and serve them in the style of a Japanese Shinkansen.

A bypass line like HS2 will just end up almost entirely empty and will be little more than a giant millstone around the neck of the railway.

HS2 wasn't a bypass line before it was hacked up by accountant-'brained' politicians. What prevents HS3 or UK Shinkansen from suffering the same? How will HS2 be an empty millstone when it connects the three largest urban areas, and spur-links to four more from the top ten. While this HS3 plan connects the 1st (London) with 11 (Bristol) and 17 (Cardiff), and maybe 26th (Newport) largest urban areas. Pre-covid, if those were timetables we want to return to, were there more 125mph super fast (non-stop or one-stop?) trains to south Wales than to Manchester?

 

[Mgameing123]

1.) In general, yes, but the major load centres could be served by dual easterly/westerly spurs.

For example, my Reading spur plan was an easterly spur following the A329(M)/A3290 into central Reading, and a westerly spur off the Berks and Hants line about half a mile east of Theale station (this is affordable, because the Berks and Hants line intersects the M4 at a location with plenty of land for a cheap and short western spur).

So trains can leave and rejoin the line at several locations to serve multiple major load centres, without too much of a time penalty)

2.) It depends how many of those new services you run through major load centres east of Bristol. Both the former county of Avon and South Wales could probably justify at least 1tph 12 car high speed service, non-stop from Heathrow as far as Newport for South Wales, or Bristol Temple Meads for former Avon.
So you could run those all the way up the HSR without touching the existing lines, as far as Barton Hill in Bristol with a Bristol Centre spur, or Magor for the South Wales services. This demand would grow over time, as people would no doubt switch from the M4/coaches with the journey time differential and new capacity depressing rail ticket prices.

Beyond that, you can do some localised widening and additional platforms at the bottleneck stations you described, without necessarily widening the railway on the most difficult sections.

All trains though should go via Reading.

== Doublepost prevention - post automatically merged: 11 Aug 2024 ==

HS2 wasn't a bypass line before it was hacked up by accountant-'brained' politicians. What prevents HS3 or UK Shinkansen from suffering the same? How will HS2 be an empty millstone when it connects the three largest urban areas, and spur-links to four more from the top ten. While this HS3 plan connects the 1st (London) with 11 (Bristol) and 17 (Cardiff), and maybe 26th (Newport) largest urban areas. Pre-covid, if those were timetables we want to return to, were there more 125mph super fast (non-stop or one-stop?) trains to south Wales than to Manchester?

If the government decided to do major cuts to HS3 then it still wouldn’t be empty if we did Paddington to Reading. That would provide the capacity needed for the Elizabeth Line, Freight and GWR semi fasts.

 

[HSTEd]

HS2 wasn't a bypass line before it was hacked up by accountant-'brained' politicians. What prevents HS3 or UK Shinkansen from suffering the same? How will HS2 be an empty millstone when it connects the three largest urban areas, and spur-links to four more from the top ten.

HS2 only achieves this if the entire project is definitely funded. Given how ambitious and all encompassing it was, this was always a very dangerous assumption to make.
As it stands, HS2 avoids the intermediate load centres between Birmingham and London, and will be (optimistically) half empty for the forseable future.

We have to design projects where they can end prematurely without becoming a complete white elephant.

While this HS3 plan connects the 1st (London) with 11 (Bristol) and 17 (Cardiff), and maybe 26th (Newport) largest urban areas. Pre-covid, if those were timetables we want to return to, were there more 125mph super fast (non-stop or one-stop?) trains to south Wales than to Manchester?

Given that there were no non-stop/one-stop trains to Manchester, that isn't really very telling!
Pre-coronavirus timetables are also a red herring in my view, that world is gone - it won't be coming back.

The line will have a capacity of 15-20 trains per hour, if you can't get close to filling that capacity you will not be able to build a business case that the government has a chance of approving.
No traffic can be left on the table here.
We are talking 15,000-20,000 seats per hour.

The only way to get near that is to take over as much of the Great Western's existing traffic as possible, which means a Shinkansen style layout with intermediate through stations.

 

[Meerkat]

Well, my plan posted before in this thread could be summarised as building a HSL following the M4 on a viaduct most of the way, but with LGV-style spurs to the existing GWML, to enable some trains to serve existing larger stations (apart from at Heathrow, which would be directly served).

So you would be able to significantly speed up longer services towards Cardiff, and increase capacity for existing load centres.

The M4 follows a winding route through areas you aren’t going to be allowed to build a viaduct.

 

[Mgameing123]

HS2 only achieves this if the entire project is definitely funded. Given how ambitious and all encompassing it was, this was always a very dangerous assumption to make.
As it stands, HS2 avoids the intermediate load centres between Birmingham and London, and will be (optimistically) half empty for the forseable future.

We have to design projects where they can end prematurely without becoming a complete white elephant.


Given that there were no non-stop/one-stop trains to Manchester, that isn't really very telling!
Pre-coronavirus timetables are also a red herring in my view, that world is gone - it won't be coming back.

The line will have a capacity of 15-20 trains per hour, if you can't get close to filling that capacity you will not be able to build a business case that the government has a chance of approving.
No traffic can be left on the table here.
We are talking 15,000-20,000 seats per hour.

The only way to get near that is to take over as much of the Great Western's existing traffic as possible, which means a Shinkansen style layout with intermediate through stations.

HS2 going nonstop between London and Birmingham is perfectly fine. As long as they actually funded the rest of the project which got scrapped because of short term thinking government. HS2 will now not be possible for a long time because no private investors will want to invest in such an unstable country.

 

[AlastairFraser]

All trains though should go via Reading.

Why? Even though I'm originally from Reading, it's not necessary for every train to stop at all. We already have several superfast services from Paddington to South Wales and the West Country which are non-stop till Bristol. There is demand for both trains that run non-stop towards their destination from London, and trains that do not.

If the government decided to do major cuts to HS3 then it still wouldn’t be empty if we did Paddington to Reading. That would provide the capacity needed for the Elizabeth Line, Freight and GWR semi fasts.

How much extra demand do you think there is for more EL services west of Hayes to justify a HSR? Most people west of Slough will take a GWR service into London if they can, as EL services are significantly slower for the same price.
GWR semi fasts - what do you do to them when you get to Reading? The new HSL services and all the other services propose can't all fit on the same tracks towards Newbury/Didcot + Oxford, Swindon etc...

With freight, this is also going to be pretty useless, as a lot of the freight services on the GWML are coming from Basingstoke (originating from Southampton port), joining the GWML west of Reading, and then leaving at Didcot. So the Paddington-Reading HSL wouldn't relieve it at all.
For any meaningful benefit, you're looking at building a HSL to Swindon, at least.

The M4 follows a winding route through areas you aren’t going to be allowed to build a viaduct.

Which areas are those?

 

[JonathanH]

Why? Even though I'm originally from Reading, it's not necessary for every train to stop at all. We already have several superfast services from Paddington to South Wales and the West Country which are non-stop till Bristol. There is demand for both trains that run non-stop towards their destination from London, and trains that do not.

We have seen the 'mistake' with HS2 that by bypassing Milton Keynes and Coventry, there remains a need for fast services on the residual route, which consume capacity that could otherwise be used for stopping services on the current fast line.

If a theoretical new line to the West doesn't stop at Reading, there still need to be fast services from Paddington to Reading, and Reading to Bristol and South Wales. Consequently you haven't moved any of the fast services off the classic route.

 

[AlastairFraser]

We have seen the 'mistake' with HS2 that by bypassing Milton Keynes and Coventry, there remains a need for fast services on the residual route, which consume capacity that could otherwise be used for stopping services on the current fast line.

If a theoretical new line to the West doesn't stop at Reading, there still need to be fast services from Paddington to Reading, and Reading to Bristol and South Wales. Consequently you haven't moved any of the fast services off the classic route.

There would still need to be fast services on the routes you have described, but less of them, and the westbound/eastbound spurs for major load centres like Reading that I have described enables services to join and leave the HSR line for calls at these stations if required.

 

[Meerkat]

Which areas are those?

Pretty much all the way from London to Reading!

 

[AlastairFraser]

Pretty much all the way from London to Reading!

There are already several large viaducts on and around the M4 alignment, some built quite recently, so I don't see the issue. If you built it away from the M4, you may be right, but a viaduct in the median wouldn't additionally detract from the environment in any of the areas the M4 passes through.

 

[Topological]

We have seen the 'mistake' with HS2 that by bypassing Milton Keynes and Coventry, there remains a need for fast services on the residual route, which consume capacity that could otherwise be used for stopping services on the current fast line.

If a theoretical new line to the West doesn't stop at Reading, there still need to be fast services from Paddington to Reading, and Reading to Bristol and South Wales. Consequently you haven't moved any of the fast services off the classic route.

Maybe one day there will be a station near the crossing with EWR. It would seem obvious if the full EWR is built. It is not quite Milton Keynes, but if plugged into the road network better it might work. Possibly pick up some of the Oxford commuters too.

I agree that we missed opportunities when planning HS2. Stops on loops do kill capacity, but with the right flighting of trains, and the acceptance that most trains only stop at one of the stops, they can be used reasonably effectively.

 

[Meerkat]

There are already several large viaducts on and around the M4 alignment, some built quite recently, so I don't see the issue. If you built it away from the M4, you may be right, but a viaduct in the median wouldn't additionally detract from the environment in any of the areas the M4 passes through.

The M4 is not straight enough for a median HS line, there probably isn’t room in the median, and the construction disruption would be monumental!
it would also have to be pretty high to get over the M25 junction.
Wouldn’t be allowed through Osterley Park, nor the Windsor/Slough/Maidenhead area.

 

[HSTEd]

There would still need to be fast services on the routes you have described, but less of them, and the westbound/eastbound spurs for major load centres like Reading that I have described enables services to join and leave the HSR line for calls at these stations if required.

The spurs could end up substantially increasing the journey time penalty for stopping at intermediate stations, since it would involve a crawl of several kilometres on the classic line to do so.
Modern high speed sets with (almost or) all axles motored can accelerate extremely rapidly, so doing anything that might cap their speed will hurt you.

Given that a non stop train from London to Cardiff would reach it in ~40 minutes, I'm not sure there would be much scope for intermediate stops using the spur approach.
The distances are simply too small - we run out of country before long non-stop runs can really be taken advantage of.

== Doublepost prevention - post automatically merged: 11 Aug 2024 ==

Why? Even though I'm originally from Reading, it's not necessary for every train to stop at all. We already have several superfast services from Paddington to South Wales and the West Country which are non-stop till Bristol. There is demand for both trains that run non-stop towards their destination from London, and trains that do not.

Can these places fill several 400m trains with 1100+ seats per hour?
I'm doubtful they can, and if they can't then running non stop is not brilliant for the railway

For context, using the rail data viewer website, the most popular destinations from London Paddington are:

Tottenham Court Road	2,197,641
Reading	2,045,987
Bond Street	1,860,320
Heathrow Terminals 2 and 3	1,671,259
London Liverpool Street	1,234,097
Farringdon	1,232,881
Oxford	1,053,224
Ealing Broadway	973,960
Heathrow Terminal 5 (Rail Station O	887,372
Slough	857,389
Hayes and Harlington	829,580
Maidenhead	727,679
London Paddington	705,117
Bristol Temple Meads	703,334
Southall	676,075
Bath Spa	593,445
Didcot Parkway	478,378
Canary Wharf	464,053
Whitechapel	438,049
Swindon (Wilts)	402,788
West Drayton	390,747
Cardiff Central	387,229

I have highlighted in italics the Crossrail destinations east of Paddington because they aren't really relevant to this discussion.
It's also worth noting the data has more people from Paddington to Paddington than to Cardiff.
Reading is the most popular destination that isn't Crossrail east of Paddington.
The next relevant destinations appear to be Heathrow and Oxford.

Cardiff and Bristol are quite a long way down the list!

From this I suggest that a high speed line heading along the GWML corridor has to serve Reading and Oxford, because that is where the traffic is.

 

[Mgameing123]

The spurs could end up substantially increasing the journey time penalty for stopping at intermediate stations, since it would involve a crawl of several kilometres on the classic line to do so.
Modern high speed sets with (almost or) all axles motored can accelerate extremely rapidly, so doing anything that might cap their speed will hurt you.

Given that a non stop train from London to Cardiff would reach it in ~40 minutes, I'm not sure there would be much scope for intermediate stops using the spur approach.
The distances are simply too small - we run out of country before long non-stop runs can really be taken advantage of.

== Doublepost prevention - post automatically merged: 11 Aug 2024 ==


Can these places fill several 400m trains with 1100+ seats per hour?
I'm doubtful they can, and if they can't then running non stop is not brilliant for the railway

For context, using the rail data viewer website, the most popular destinations from London Paddington are:

Tottenham Court Road	2,197,641
Reading	2,045,987
Bond Street	1,860,320
Heathrow Terminals 2 and 3	1,671,259
London Liverpool Street	1,234,097
Farringdon	1,232,881
Oxford	1,053,224
Ealing Broadway	973,960
Heathrow Terminal 5 (Rail Station O	887,372
Slough	857,389
Hayes and Harlington	829,580
Maidenhead	727,679
London Paddington	705,117
Bristol Temple Meads	703,334
Southall	676,075
Bath Spa	593,445
Didcot Parkway	478,378
Canary Wharf	464,053
Whitechapel	438,049
Swindon (Wilts)	402,788
West Drayton	390,747
Cardiff Central	387,229

I have highlighted in italics the Crossrail destinations east of Paddington because they aren't really relevant to this discussion.
It's also worth noting the data has more people from Paddington to Paddington than to Cardiff.
Reading is the most popular destination that isn't Crossrail east of Paddington.
The next relevant destinations appear to be Heathrow and Oxford.

Cardiff and Bristol are quite a long way down the list!

From this I suggest that a high speed line heading along the GWML corridor has to serve Reading and Oxford, because that is where the traffic is.

A non stop would definitely not work with 400m trains but could be achieved by using the Carmathen services.

 

[Topological]

Doing the best Crayonista...

IF you built the Heathrow to Colne Valley HS2 section as a first step

Then build from Colne Valley towards Wycombe and Oxford roughly along the M40, then swing South West to Swindon along the A420, expand the line through Swindon to accommodate HS tracks as well as the existing, then upgrade Swindon to Bristol Parkway that would get you the South West and Oxford on the same line. Misses Reading, but frees the GWR fasts for more Reading capacity (the trains will not arrive at Reading full).

The junctions in Colne Valley allow North West to Heathrow and South West to OOC/Euston.

The M40 and A420 are only indicative of the route. I would actually expect Oxford to be "Oxford South" with the line running across the bottom of Oxford.

 

[AlastairFraser]

The M4 is not straight enough for a median HS line, there probably isn’t room in the median, and the construction disruption would be monumental!
it would also have to be pretty high to get over the M25 junction.
Wouldn’t be allowed through Osterley Park, nor the Windsor/Slough/Maidenhead area.

1.) The pillars for the viaduct would be in the median as much as possible - I accept this won't be possible everywhere.

2.) If you check my earlier proposal, my tunnel through the Heathrow station starts at Heston services and ends at M4 Junction 5 in Langley, so travelling over the M25 isn't an issue here.

3.) With Osterley Park, it's possible to extend my Central London to M4 tunnel a mile west to bypass it easily. I'm not convinced that a viaduct would be particularly visible from Osterley Park house, given that the treeline on the M4 boundary is quite high, but the option is there with.

With the Windsor area, it's not a given that there will be issues. There are several existing large viaducts or bridges in the area that carry roads or other transport infrastructure (the Queen Elizabeth Bridge in Windsor and the A332/A355 viaduct continuations into Slough, the M4 bridge over the Thames in Dorney/Bray, to name a couple). This viaduct will be following an existing transport corridor, and will benefit a lot of local residents immensely, with the removal of a lot of congestion.

The spurs could end up substantially increasing the journey time penalty for stopping at intermediate stations, since it would involve a crawl of several kilometres on the classic line to do so.
Modern high speed sets with (almost or) all axles motored can accelerate extremely rapidly, so doing anything that might cap their speed will hurt you.

Given that a non stop train from London to Cardiff would reach it in ~40 minutes, I'm not sure there would be much scope for intermediate stops using the spur approach.
The distances are simply too small - we run out of country before long non-stop runs can really be taken advantage of.

== Doublepost prevention - post automatically merged: 11 Aug 2024 ==


Can these places fill several 400m trains with 1100+ seats per hour?
I'm doubtful they can, and if they can't then running non stop is not brilliant for the railway

For context, using the rail data viewer website, the most popular destinations from London Paddington are:

Tottenham Court Road	2,197,641
Reading	2,045,987
Bond Street	1,860,320
Heathrow Terminals 2 and 3	1,671,259
London Liverpool Street	1,234,097
Farringdon	1,232,881
Oxford	1,053,224
Ealing Broadway	973,960
Heathrow Terminal 5 (Rail Station O	887,372
Slough	857,389
Hayes and Harlington	829,580
Maidenhead	727,679
London Paddington	705,117
Bristol Temple Meads	703,334
Southall	676,075
Bath Spa	593,445
Didcot Parkway	478,378
Canary Wharf	464,053
Whitechapel	438,049
Swindon (Wilts)	402,788
West Drayton	390,747
Cardiff Central	387,229

I have highlighted in italics the Crossrail destinations east of Paddington because they aren't really relevant to this discussion.
It's also worth noting the data has more people from Paddington to Paddington than to Cardiff.
Reading is the most popular destination that isn't Crossrail east of Paddington.
The next relevant destinations appear to be Heathrow and Oxford.

Cardiff and Bristol are quite a long way down the list!

From this I suggest that a high speed line heading along the GWML corridor has to serve Reading and Oxford, because that is where the traffic is.

I'll address your first point here:

Let me give an example to show you that quicker intermediate stops on longer runs using spurs are possible. So westbound HSR trains stopping at Reading would leave the mainline at Winnersh Interchange on the M4, and follow a HSL standard spur on the A329(M)/A3290 to the mainline at Suttons Seeds/west end of Thames Valley Park, then into Reading station.

Approx. 4.84 miles of the 6.07 miles from the main HS3 line at Winnersh Interchange will be on HSL, leaving a mile and a quarter into the station. Is this going to have a massive impact on deceleration?

I admit the first section of the route west out of Reading would be lower speed, with the 60 limit at Southcote Junction, then 100 up to the westbound spur at the M4 overpass near Calcot and Theale. It would be approx. 5 miles from Reading station back up to the HS3 mainline at the approaches for M4 J12.

Reading to Theale station currently takes 6 mins on a Class 387, with the superior acceleration and slightly shorter distance to the HS3 mainline, this could be cut to 5 mins. Is around 15 mins for the stop at Reading and return to the HS3 mainline really a massive issue?

As to your second point, I think it is prudent to point out that current passenger volumes to more far flung destinations are suppressed by price and time taken to reach key destinations.

For example, Bristol currently has an average of 55 coaches a day to London (according to Trainline), and even half of that total coach capacity would be an extra 1500 passengers a day. The Annual Average Daily Traffic (AADT) for sections of the M4 east of Bristol is around 30k a day, a large portion of that will be heading east towards the Thames Valley and London.

Viability of a HSR line has to rest on significant abstraction from other sources, and this would be no exception.

A HSR line through both Reading and Oxford towards the West Country and South Wales is impossible without taking an long detour. If you want additional capacity towards Oxford, the obvious and cheaper thing to do (as well as HS3 along the M4 corridor) would be reinstatement of 4 track, speed upgrades and electrification of the entire Chiltern Main Line, and the branch from Bicester to Oxford.
Currently it is 1hr14 Marylebone to Oxford at fastest, around 23 mins slower than GWR, but you can close most of the gap with the upgrades I described.

More frequent services, including expresses, could then be run from Marylebone to Oxford along that route, without compromising the viability of the HS3 route.